2.
3. On the last point, observers have noted that goats may show extreme fright and hiding responses to aircraft (Ballard 1975; Hazelwood 1983; Jones 1984), especially from helicopters.
Mountain goats are seasonally migratory. In summer, coastal goats in the Olympic Mountains occupy areas above 1500 m (Houston et al. 1986). They use winter ranges above 1500 m and descend to rock outcrops as low as 300 m. Seasonal ranges may overlap or be up to 8 km apart. Smith and Bovee (1984) considered year-round habitat for coastal goats in Alaska to include all areas above 800 m and lower elevation winter range. Both forested habitats well below timber-line and alpine zones support high density coastal and interior goat populations in winter (see Smith and Bovee 1984). Slopes with southern aspects are preferred.
Nursery groups of females and juveniles range in size from two to 25 or more individuals. From early summer through August, group size is large on feeding grounds or at mineral licks. As summer progresses, goats disperse and groups get smaller.
Habitat types can be used to define the areas in which goats are likely to be found. Boundaries of sample units for goats should follow creeks or other low areas below the areas of goat concentration. The tree-line may set a useful lower elevation boundary in northern populations (Northern Boreal Mountains ecoprovince). If units are too large, there is risk of double counting mobile goats. Van Drimmelen (1986) used sample units from 11 to 32 km
2, with a mean of 19 km
2. Smith and Bovee (1984) reported high densities of coastal goats (2.3 goats/km
2) in year-round habitats compared to estimates in Yoho National Park (1.5 goats/km
2). According to Warkentin (pers. comm.), treeline is not necessarily a useful lower elevation boundary for goat surveys, but a temperature range of +10 ° C may be. Goats in the Kootenays are often found in a 300 m elevation band about that temperature from mid-July through early September.
4.2.3 Survey timing
Goats may be most accurately classified to sex and age during the first three weeks of July, with the best period about mid-month. At that time, identification of adult males and females is assisted by moulting patterns. Winter surveys, unless to determine seasonal distribution, are not recommended because of difficulties related both to habitat selection and the cryptic aspects of white animals on a white background. There is no seasonal timing that provides complete data on both composition and numbers. Total counts in summer focus on composition, but it is not known if the composition data are biased. The timing to provide the highest total counts may vary from place to place. Opinions on best timing include "mid-July to mid-August" (Nichols 1985), with “mid-August to late September” for coastal Alaska (Smith 1984); late August to mid-September” in the northern interior of British Columbia (Northern Boreal Mountains, Taiga Plains, Boreal Plains ecoprovinces) ; “August to early October” for southern and central British Columbia (Sub-Boreal Interior, Central Interior, Southern Interior, Coast and Mountains ecoprovinces); and “August to September” for the Southern Interior Mountains.
Early morning and late evening are poor for surveys on clear days, because of glare and shadow conditions caused by direct, low angle sunlight. The best survey conditions in terms of observer comfort, visibility, and animal activity are thought to be "...high overcast skies, soft light, no turbulence" in Alaska (Nichols 1980). Fox (1977) found goats congregating into larger groups, and moving into open, up-slope areas on clear days following stormy weather. He believed that the best counts could be accomplished in the late afternoons of such days. On clear, hot summer days, between about 11:00 and 16:00 hours, goats may remain inconspicuously bedded in cool areas such as timber, caves, crevices, north-slope shadows, ravines, and snow banks (Ballard 1975; Nichols 1980; Jury 1984). To maximize sightability, goat surveyors should keep a careful eye on prevailing weather.
4.2.4 In-flight survey procedures
Piper supercub aircraft have been used for goat surveys (Ballard 1975; Nichols 1980). However, helicopters provide greater safety, better visibility, and more flexibility in aerial maneuvering or herding animals about. Turbine driven helicopters, especially the Bell 206 with bubble windows at the rear seats are preferred, although the Hughes 500D disturbs the animals less.
Within each sample unit, flight lines should be routed over potential goat habitats in a pattern, determined in-flight, that provides opportunity to see all animals present. The best pattern is usually along contours, starting from the lowest level and working systematically upward, in jumps of 100 to 150 m depending on topography and cover. Airspeed may be varied to insure optimum sightability of goats as terrain varies.
Where long expanses of habitat require passes at several levels of elevation, coverage should be accomplished in segments bounded by recognizable natural features, such as between two glaciers or within a particular basin. If goats are observed above any contour line transect being followed, they can be noted on the map, but not formally recorded until a transect passes their elevation. That procedure ensures that moving animals are neither missed nor counted twice.
4.2.5 Classification
Ideally, goats should be directed up open mountain slopes. Goats appear to respond better to aircraft approaching from downslope, rather than overhead. Herding goats can assist in classification. Restricting goat access to escape terrain allows better classification, but can also result in pandemonium and unwanted stress. Large, dispersed groups of animals are sometimes most easily segregated as adults and kids (Level 1) during a high elevation approach. Large herds can be approached and classified at a distance, which results in less movement and stress on animals. For better viewing, large groups can be "broken" into sections and classified in smaller groups. Smaller groups can be dealt with closer at hand.
Level 3 classification should be used in July and, to the extent possible, Level 2 later. Yearlings are not readily distinguished from other young adults by late summer, especially in large groups. It is also difficult to separate young males from young females at that time, and male ratios will therefore be low. Distinguishing sub-adults (or 2 years olds - Level 4) can be difficult. Hatler and Hazelwood (1985) expressed some doubt about making that distinction from the air, especially in large groups, and a recent study in Montana (Smith 1988) has confirmed that difficulty.
Moult and body size have helped classification of goats in July (Nichols 1980; Hatler and Hazelwood 1984; 1985). Horn morphology, commonly cited as an aid in classification, is difficult to apply reliably from the air. The flow of decisions in classification in Figure 1 was adapted from Hatler and Hazelwood (1984). It should be confirmed with the features from
Figure 1. Flow of decision making in classification of goats (adapted from Hatler and Hazelwood 1984).
Hatler and Hazelwood (1984) noted that most classification errors occur in the smooth pelage category. Young males are easily missed in large groups, and sub-adults cannot be readily distinguished from adult females. Shaggy nannies and yearlings can also be confused in large groups. Dry nannies will moult earlier than those bearing young. Kids can often be missed because they hide behind or under nannies.
After July, moult features no longer help in classification, and the genitalia are usually obscured by lengthening body hair. Other features still apply, but it may not be possible to sex adults with confidence. Juveniles are still recognizable by body size, but yearlings have grown to sizes overlapping with older classes, and cannot be consistently segregated. Thus, a somewhat compromised Level 2 is the only practical classification standard for surveys in late summer and fall.
4.2.6 Estimating numbers missed
Few mountain goats have been radio-collared in British Columbia, so estimates of sightability using mark/resight analysis have not been possible. Many surveys have used the assumption that sightability is not biased between sex and age classes, and that all goats are seen (Smith and Bovee 1984). However, Foster (1982) reported average sightability of only 42% in west-central British Columbia (Central Interior ecoprovince). Smith (1984) reported 30% sightability in coastal Alaska, using fixed-wing aircraft and telemetry. Warkentin (pers. comm.) suggests that: sightability of kids is lower in escape terrain as they hide behind nannies; sightability of billies is lower than nannies from mid-July through August; and sightability of nanny and kid groups is lower than that of billies in June and early July, when they may still be on mineral licks in timber.
Smith and Bovee (1984), using mark/resight surveys on a coastal goat population, calculated density estimates on winter and on year-round range at 4.4 and 2.0 goats/km
2 respectively. They reported an estimate for goats, varying +15% about the mean at confidence levels of p=0.95. Because of the "limited and variable sightability of coastal goats, they thought that no uncorrected count should be taken as more than an index of the population size".
Cichowski et al. (1991) used mark/resight techniques to study goat populations in the Babine Mountains Recreation Area. Goats were marked from a Bell 206 helicopter using a paint gun. One goat was marked per 5 minutes of flying time. Observing goats on both sides for paint marks proved difficult in large groups (>20). Cichowski et al. (1991) recommend marking all goats on the same side to increase survey efficiency. It is not known if marked and unmarked goats were equally observable.
[Back to TOC] [Previous] [Next]
Copyright © 1997 Province of British Columbia
Published by the Resources Inventory Committee