Mineral Lick Visitation by Mountain Goats, Oreamnos americanus

CLIFFORD G. RICE
Wildlife Program, Washington Department of Fish and Wildlife, 600 Capitol Way N., Olympia, Washington 98501 USA; e-mail:
ricecgr@dfw.wa.gov.

Rice, Clifford G. 2010. Mineral lick visitation by Mountain Goats, Oreamnos americanus. Canadian Field-Naturalist
124(3): 225–237.
Many species, including Mountain Goats (Oreamnos americanus), are known to visit mineral licks, but the extent and dura-
tion of use are poorly understood because most studies consist of observations at licks. I studied the movements to, from,
and near mineral licks of 11 mountain goats in Washington wearing Global Positioning System (GPS) collars for a total of
169 goat-months of tracking and evaluated chemical composition of six mineral licks compared with reference soil samples.
I recorded 101 mineral lick visits to 13 mineral licks. Each GPS fix was classified as moving toward a mineral lick, in the
vicinity of a lick, on an excursion from a lick, moving away from a lick, or not associated with lick use. Depending on
annual movement patterns associated with lick use, each Mountain Goat was classified as a Migrant (single lick visit of
long duration, n = 3 Mountain Goats), Sojourner (few visits of short duration, n = 2), Commuter (many visits of short dura-
tion, n = 5), or Resident (lick within normal range of movements, n = 1). Most mineral lick visits took place 01 June–15
August with peak visitation about 14 June–29 July. Migrants typically stayed in the vicinity of licks about a month (but as
long as 51 days) whereas other mountain goats visited licks for 0.1–8 days (median = 1 day). Migrants also tended to take
longer and move farther than other Mountain Goats when on movements to and from licks. Most Mountain Goats moved
toward mineral licks faster (km/hr) than they moved away from licks. All licks had higher concentrations of sodium than
reference samples (1.5–27 times as high), although concentrations of calcium, potassium, and sulphate tended to be higher
as well, whereas magnesium was not. Mineral lick visitation has costs (energetics of travel, reduced forage, and predation
risk). Depending on the importance of these costs, mountain goats evidently use various strategies for exploiting mineral
licks as exemplified by the movement types (migrant, sojourner, commuter, or resident). Notably, most of the Mountain
Goats in this study crossed national forest, county and Washington Department of Fish and Wildlife region boundaries to
another to visit mineral licks. Thus, coordination among administrative units is needed in management of Mountain Goats
and mineral licks they use.
Key Words: Mountain Goats, Oreamnos americanus, Global Positioning System (GPS), mineral lick, movements, Washington.

The use of mineral licks has been documented for erence soil samples and were deficient in forage (Klein
numerous species (Klaus and Schmid 1998), especial- and Thing 1989; Ayotte et al. 2006; Mincher et al.
ly ungulates (Jones and Hanson 1985; Kreulen 1985; 2008). Also, during the time of greatest lick use (late
and many others) including mountain goats (Oream- spring and early summer), sodium requirements are
nos americanus, e.g., Hebert and Cowan 1971; Singer high as this corresponds to late gestation and early lac-
and Doherty 1985; Poole and Heard 2003; Turney and tation for many species and elevated potassium in for-
Blume 2004; Poole et al. 2010). Most studies have age plants at that time increases the need for sodium
emphasized chemical composition of lick soils (e.g., (Weeks and Kirkpatrick 1976; Atwood and Weeks
Kennedy et al. 1995; Tracy and McNaughton 1995, 2002). Foley et al. (1995) indicated that sodium may
Dormaar and Walker 1996; Ayotte et al. 2006) or mon- be important in mediating the acidification resulting
itored visitation to licks (e.g., Tankersley 1984; Moe from detoxification of forage secondary compounds.
1993; Atwood and Weeks 2002; Turney and Blume Others have suggested that magnesium is a key com-
2004) and provide at most, anecdotal accounts of ponent of lick soils (Jones and Hanson 1985; Heimer
movements to and from mineral licks (Heimer 1974; 1988; Klaus and Schmid 1998).
Tankersley 1984; Hnilicka et al. 2002). One exception Despite these evident benefits, visiting mineral licks
to this is the study of movements of White-tailed Deer incurs costs. These are in the form of energetic costs
(Odocoileus virginianus) visiting licks in Indiana (Wiles of movement, potentially increased risk of predation
and Weeks 1986). in unfamiliar terrain and suboptimal habitat lacking
Although there has been considerable discussion of escape terrain, and loss of foraging opportunity due to
the potential benefit from mineral licks (Kreulen 1985; time budget constraints and potentially reduced for-
Klaus and Schmid 1998), for ungulates in seasonal age availability due to poor habitat or high levels of
environments, the preponderance of evidence points to use in the vicinity of the lick. Because the costs and
sodium as the constituent primarily associated with benefits vary among licks and among individuals,
licks (Kennedy et al. 1995; Klaus and Schmid 1998; patterns of mineral lick use can be expected to vary
Atwood and Weeks 2002). Sodium concentrations in depending on the quality of the lick, distance to the
lick soils were consistently much higher than in ref- lick, terrain that must be traversed to visit the lick,

225
226 THE CANADIAN FIELD-NATURALIST Vol. 124

mineral content of the soils and vegetation in the “nor- mineral licks was an adult male (038GMM) and the
mal” home range of the individual, and habitat in the extent to which his mineral lick use reflected his sex
vicinity of the lick. (Hebert and Cowan 1971; Ayotte et al. 2008; Poole et
Prior to this study, knowledge of mineral lick use by al. 2010) or that he had a mineral lick within his nor-
Mountain Goats in Washington was limited. Wright mal range is an open question. 051GPM was also a
(1977) reported Mountain Goats using a mineral lick male, but was captured at 1 year of age, and, in terms
on the flanks of Mt. Baker (exact location not speci- of lick visitation, probably behaved more like the adult
fied). Representatives of the Sauk-Suiattle Tribe deter- females during the 13 months he was tracked.
mined that the early summer aggregation of mountain Most Mountain Goats in this study captured prior
goats on Gamma Ridge (Glacier Peak) was associated to 2006 (033GPF, 034GPF, 037HRF, 045MRF, and
with the use of a mineral lick (D. Graupman personal 038GMM) were set on a 3 hr fix interval. The excep-
communication 2001). To increase our understanding tion was 024KRF which was on a 5 hr rotating sched-
of the use of mineral licks by Mountain Goats in ule from 13 September 2003 to 28 June 2004, and sub-
Washington, the objectives in this study were to iden- sequently on a 3 hr interval. The collars for the 2006
tify mineral licks used by Mountain Goats, and to des- captures (051GPM, 052GPF, 053GPF, 054GPF, and
cribe movements of GPS collared Mountain Goats vis- 055GPF) were on a 5 hr rotating schedule most of the
iting mineral licks. Because there are few comparable year but on a 1 hr interval 07–21 June and 25 July–20
studies of mineral lick visitation by Mountain Goats, August, when movements to and from Gamma Ridge
I described the frequency with which individuals vis- were expected. I removed outlier fixes by visually
it mineral licks, the distances they travel in doing so, screening locations beyond the continuous distribu-
to further our understanding of how they balance the tion of distances of all fixes from the median for each
trade-offs between costs and benefits of mineral lick individual and by visual inspection of travel paths (usu-
use. I also evaluated the chemical constituents of min- ally single fixes separate from temporal clusters, Rice
eral licks mountain goats used. 2008). The median estimated location error (Lewis et
al. 2007) for fixes used to determine movements to
Materials and Methods mineral licks and habitat was 9.6 m (central 95% =
Study Area 5.9–129.8, n = 9165).
I studied Mountain Goat mineral licks in the Cas-
cade Range in Washington State where topography Identifying Licks
extends as high as 4267 m on several volcanic peaks, Mineral licks used by collared Mountain Goats
but most terrain is below 2100 m. In this area, Moun- were identified in three ways: on the basis of previous
tain Goats typically inhabit elevations between 600 and knowledge; field observations; and in two cases, move-
2400 m, which have six broad classes of vegetative ment records from GPS collars which were to clusters
cover (derived from Comer et al. 2003): Bare (1510– of fixes 4-5 km outside the range of movements for
4270 m), which includes bedrock, scree, talus, and that individual. Other mineral licks probably occurred
dwarf shrubs; Grassland (1170–2190 m); Parkland within areas of movement for some collared mountain
(1180–2080 m); Woodland (600–1840 m); Forest goats. Such licks might or might not be associated with
(600– 1470 m); and Shrubland (600–1380 m). clusters of collar fixes as this was variable for known
licks. Within the usual movements for an individual, it
Capture and Collaring was not feasible to distinguish clusters of fixes associ-
I captured 46 Mountain Goats in the Cascade Moun- ated with licks from those associated with other loca-
tain Range between 26 September 2002 and 2 July tions of high use (e.g., favored resting sites) without
2007 (Rice and Hall 2007) and fitted them with GPS independent information on the location of the lick.
tracking collars (Vectronic GPS Plus-4, Vectronics Field observations were made on mineral lick use on
Aerospace, Berlin, Germany). All captures were in
Gamma Ridge on 25 and 26 July 2007, during which
compliance with Washington Department of Fish and
six new mineral lick sites were identified by observ-
Wildlife Policy on Wildlife Restraint or Immobiliza-
ing mountain goat use.
tion (M6003). The primary purpose of these captures
was to provide locations for studies of movements and Movements
habitat selection (Wells 2006) and to provide marked I identified five states for mountain goats deter-
groups for sightability modeling for surveys (Rice et al. mined by movements relative to mineral licks: (1) At
2009). Only the 11 collared mountain goats known to Lick, (2) moving Toward the lick, (3) moving Away
visit mineral licks are included in this study (Table 1). from the lick, (4) on an Excursion from the lick, and
Seven of these were captured near the licks on Gamma (5) None (none of the above), collectively termed Lick
Ridge (Glacier Peak) in 2006 to improve our under- State. A Mountain Goat was considered At Lick if the
standing of movements of Mountain Goats associated fix was within the vicinity and within a specified dis-
with mineral licks and Gamma Ridge in particular. tance from the lick (same as region 2 of Hebert and
One capture took place in the vicinity of the Deadhorse Cowan 1971). This distance varied among the Moun-
Point lick. Because our overall study emphasized fe- tain Goats and was determined by examining time
males, only 1 of the 11 mountain goats that visited series graphs of distance from the lick and between-
2010 RICE: LICK VISITATION BY MOUNTAIN GOATS 227

TABLE 1. Mountain Goats visiting mineral licks tracked with GPS collars in Washington. Sex is indicated by the last letter of the individual designation (F = female, M = male). Age is fix paths for each lick area. Generally, starting and
ending fixes of movements Toward and Away were

distance
Max

1400
1400
2475
400
1275
850
700
1150
1050
850
625
clearly evident in time series graphs of distance from
the lick, but when questions arose, I used the rule that
the movements were considered continuous if the dis-
tances for the fixes in question covered a period of
< 2 days. Excursions were movements Away immedi-
ately followed by movements Toward which did not
Visits
1
5
2
5

101
7
11
4
3
13
5
45
reach the typical distances of Toward and Away for
that individual.
Based on the Lick States, I calculated the Duration
of each State, and the change in Distance to the lick (in
km) between the first and last fixes in each State. I
Months
tracked
11
26
13
26

169
19
11
2
11
15
8
26
defined the Interval between mineral lick visits as the
difference (in days) between the start of movement
Toward and the end of the previous movement Away
for each visit and partitioned these Intervals into those
that were within a given year and those between years
(over winter).
Fixes
2036
3200
2025
4711

28605
3055
1841
780
1549
3067
1319
5022

Based on movement records, I classified each indi-

vidual as one of four Types: Migrant, Commuter,
Sojourner, or Resident. Migrants moved to the lick
and stayed for an extended period (>2 weeks). Com-
muters moved to and from the lick frequently within a
6 June 2005
22 September 2006
25 July 2007
14 November 2005

19 September 2007
10 February 2006
10 August 2005
5 September 2006
30 May 2007
19 September 2007
14 March 2007
9 September 2006

season. Sojourners visited the licks briefly, and if they

at time of capture. Max distance is the maximum distance from the nearest mineral lick when At Lick (m).

visited a lick more than once in a season, visits were

Maximum

separated by >2 weeks. The Resident visited a lick

located within his normal range of movements.
Analysis
To test for differences in measures of mineral lick
use among visitor Types and Lick States, I included
Dates

individual identity as a random effect because multi-

ple visits are repeated measures on the same individ-
ual and used the Tukey test for multiple comparisons
9 July 2004
9 July 2004
29 June 2006
13 September 2003

13 September 2003
10 July 2004
3 September 2004
29 June 2006
29 June 2006
29 June 2006
30 June 2006
10 July 2004
Minimum

of the means of the different groups (Zar 1996). To ad-

dress distribution considerations (skewness of pooled
samples = 0.813–4.121), I log-transformed Durations,
Intervals, Distances and movement rates. For move-
ment rate analysis, I adjusted for the fact that collars
were programmed with varying fix intervals and that
not all fix attempts were successful by including real-
French Cabin Creek

ized fix interval (in hours) as a categorical nuisance

Gardner Mountain
Mineral Mountain

variable. Statistical analysis was conducted with JMP

Deadhorse Point
Gamma Ridge
Gamma Ridge
Gamma Ridge

Gamma Ridge
Gamma Ridge
Gamma Ridge
Gamma Ridge

(v7.0, SAS Institute 2007).

Lick Sampling and Analysis
Soil samples were collected at two mineral licks on
Area

Gardner Mountain and five licks on Gamma Ridge

(Figure 1). At each site, reference samples were col-
lected 50 m upslope, downslope, and to each side of
Age

the site. Each sample was analyzed for chemical con-

4
4
1
3
3
5
6
5
4
2
3

stituents frequently referenced in the earlier reports

038GMM
051GPM

Commuter 045MRF
Sojourner 024KRF
037HRF
033GPF
034GPF

052GPF
053GPF
054GPF
055GPF

(e.g., Kennedy et al. 1995; Ayotte et al. 2006): Sodium

Goat

(Na), Calcium (Ca), Potassium (K), Magnesium (Mg),

All

and sulphate (SO4) which were assayed by Kuo Soil

Labs (Othello, Washington). Because the distributions
Resident

of chemical concentrations were skewed, I log-trans-

Migrant

formed all values. Te evaluate if chemical concentra-

Type

tions differed between lick and reverence samples, I

228 THE CANADIAN FIELD-NATURALIST Vol. 124

used a nested ANOVA design (lick vs reference within (28 and 21), and lowest for Sojourners (13).
site). I also checked that reference samples from the Migrants typically stayed At Lick >1 month per
downslope of the site did not differ from other refer- visit (Table 2), which was significantly longer than oth-
ence samples with a 1- sided t-test, in case drainage er Types (Table 2, F3,6 = 10.251, P = 0.009). Sojourners
from the lick site may have elevated concentrations were usually At Lick 2–3 days and not more than about
for the downslope sample. one week whereas Commuters usually stayed at lick
about one day or one week at the most. The Resident’s
Results visits were shorter (Table 2), although differences
Movements among Sojourners, Commuters and Resident were not
Of the eight Mountain Goats captured near mineral significant. The Duration of movements Toward were
licks, three were Migrants, one was a Sojourner, and less than Away (Table 2, F1,124 = 9.793, P = 0.002,
four were Commuters. The Migrants stayed in the Table 2).
vicinity of the lick a median of 38.6 (range 23.9-38.7) Within a given year, Commuters and the Resident
days after capture. The Sojourner stayed 5.9 after cap- had similar Interval between mineral lick visits, which
ture, and the Commuters stayed in the vicinity a medi- was much shorter than those for Sojourners (F2,49 =
an of 2.8 (range 2.6-3.5) days after capture. It would 5.816, P = 0.005, Table 3). Between year Interval
appear that capture did not have much effect on lick observations were too sparse for meaningful testing
visitation (see below). (Table 3).
Mountain Goats generally followed mountain ridges The longest Distance a Mountain Goat moved
when moving Toward and Away from mineral licks Toward or Away from a mineral lick was 29.4 km and
(Figure 2). However, this was only partially true for the shortest Distance was 0.6 km (Table 4). Typical
Migrants, which followed ridges initially when mov- movements for Migrants were >15 km, whereas those
ing toward the licks, but then crossed the Suiattle River for Sojourners and Commuters were more variable (4–
valley rather than detour along the ridge to the south 17 km), with the Resident’s movements usually much
(paths in the upper half of box for Gamma Ridge, Fig- shorter although these differences were marginally not
ure 2). Also, 045MRF crossed the lower part of the significant (Away F3,6 = 4.526, P = 0.059; Toward
Winthrop Glacier on her many trips Toward and Away. (F3,5 = 5.409, P = 0.050). Generally, Distance moved
Movements Toward and Away from mineral licks by was slightly greater Toward than Away (F1,122 = 6.245,
Mountain Goats were usually decisive (Figure 3), but P = 0.014).), movement rate (m/hr) was highest dur-
there were exceptions. For instance, in 2006, 034GPF ing Toward, nearly three times the rate when in lick
stopped her movement 6 km from the Gamma Ridge state None and Away was also higher than lick state
mineral licks, retreated to 11 km from the licks for None (Table 5, F4, 8395 = 101.941, P < 0.001). This
four days, resumed her movement toward the lick, but demonstrates the energetic costs of mineral lick move-
paused again at 6 km for 2.5 days before moving to the ments. Movement rates did not vary significantly
licks (Figure 3). In 2005, 034GPF turned and ascended among Types (F3, 7 = 0.492, P = 0.699).
along the Suiattle River 4 km over two days before
continuing to the licks. Although the mineral licks for Soil Samples
the Resident (038GMM) was enclosed by other areas Sodium concentrations were significantly higher in
he visited, the radial nature of his Toward and Away lick soils than in reference soils at all licks and were
moments suggests that the lick was the main reason >20 times higher for two licks (GAM1 and GAM3,
he visited the area of the lick (Figure 2). Table 6). Other chemicals had significantly higher
Most (90%) mineral lick visits took place 1 June– concentrations at some licks (K at 3 licks, Mg at 1, and
15 August but this varied among Types. Apart from a SO4 at 1). Despite the lack of consistently significant
few early visits to mineral licks, Migrants, Sojourners, differences, concentrations of Ca, K, Mg, and SO4
and Commuters, all started lick visits in mid-June (14 were generally higher at licks than in reference sam-
June, 14 June, and 17 June, respectively, Figure 4). The ples. Downslope samples were not significantly higher
Resident started regular lick visits on 29 April, and than the other reference samples at each site for any
activity increased on 25 May (Figure 4). All Types chemicals with regards to either concentration of dif-
ceased regular visits near the end of August (Migrants- ference from the site sample (t1–2 = 0.016–0.984, 3 of
24 August, Sojourners-19 August, Commuters-26 70 tests were with P < 0.05).
August, and Resident 21 August) although there were
gaps in visitation for Sojourners (13 July–02 August), Discussion
Commuters (09–17 August), and the Resident (26 We tracked 46 Mountain Goats distributed over a
July–11 August). Altogether, peak visitation was about wide geographic range in Washington (Rice and Hall
14 June–29 July. Early and late visits to mineral licks 2007; Rice 2008), but only 11 exhibited pronounced
also occurred, especially for the Resident, but also for movements associated with mineral licks and eight of
other Types except Migrants (Figure 4). Typical num- these were captured while visiting known licks. The
ber of days At Lick in a year was highest for Migrants results show that there is wide variation in the details
(44), intermediate for the Resident and Commuters of mineral lick visitation among individuals in terms
2010 RICE: LICK VISITATION BY MOUNTAIN GOATS 229

FIGURE 1. Examples of Mountain Goat mineral licks in Washington, 13 September 2003–19 September 2007.

of number of visits, distance traveled, and time spent Those Mountain Goats that visited mineral licks did
in the vicinity of licks leading to the categorization of so every year they were tracked. Poole et al. (2010)
lick visitation into four types. As with other species, assert that most populations of Mountain Goats make
lick use was decidedly seasonal for Mountain Goats. extensive use of natural licks and detected extra-range
None of the licks we visited were at the bases trees, in lick visitation in about 70% of their collared Mountain
contrast to those investigated by Poole et al. (2010) in Goats in two populations. They also noted the diffi-
southeastern British Columbia. culty in documenting lick use from GPS collar records
230 THE CANADIAN FIELD-NATURALIST Vol. 124

FIGURE 2. Movements of GPS-collared mountain goats toward and away from mineral licks in Washington, 13 September
2003–19 September 2007. Non-lick paths for each Mountain Goat are indicated by callout boxes giving the goat
name and Type in parenthesis (M = Migrant, S = Sojourner, C = Commuter, and R = Resident).
2010 RICE: LICK VISITATION BY MOUNTAIN GOATS 231

FIGURE 3. Distance from their respective mineral licks for 11 Mountain Goats in Washington, 13 September 2003–19 September
2007. Multiple lines for an individual indicate multiple years of tracking.
 232
TABLE 2. Median, maximum, and minimum Duration of Lick States (moving Toward lick, At Lick, Excursion from lick, and moving Away from lick) in days for individual Mountain THE CANADIAN FIELD-NATURALIST Vol. 124

within a Mountain Goat’s normal range of movements.

1 Thus, just how many of the 35 mountain goats we
3
1
5
5
3
8
10
4
2
12
3
31
26
70
n

tracked for which we could not document mineral lick

use did not use licks and how many used licks we did
Max
6.3
7.0
1.0
7.0
2.8
3.6
3.6
1.6
6.3
4.9
4.6
2.0
6.3
4.9
7.0
not detect remains an open question, but my subjective
Away

assessment is that a number of them did not use licks.

Min
6.3
2.4
1.0
1.0
1.5
1.9
1.5
0.3
1.7
2.5
0.5
1.3
0.3
0.1
0.1
Notably, Festa-Bianchet and Côté (2008) did not report
any natural mineral lick use over 15 years of study of
the Caw Ridge (Alberta) Mountain Goat population
Median
6.3
5.1
1.0
5.1
2.1
3.1
2.1
1.0
3.1
3.7
1.7
1.5
1.5
1.0
1.5
and Fox et al. (1989) indicated that lick use was not
evident in Mountain Goats in southeast Alaska. It may
be that the generally high rates of use of mineral licks
by Mountain Goat populations is a consequence of the
availability of licks in mountainous terrain and the
0
2
0
2
0
4
4
1
0
1
1
1
4
19
29
n

lack of lick use is due to the lack of availability.

The seasonal nature of mineral lick use has been
Max

2.5

2.5

0.3
0.3
0.8

2.3
0.2
1.7
0.2
0.1
0.1

reported in other species (e.g. Weeks and Kirkpatrick

.

.
Excursion

1976; Tankersley 1984; Atwood and Weeks 2002) and

Mountain Goats (Hebert and Cowan 1971; Turney and
Min

5.1

5.1

3.0
3.0
0.8

2.3
0.2
1.7
2.3
2.3
5.1
.

Blume 2004). For mountain ungulates, the period of

greatest visitation was similar to those of this study
Median

(Heimer 1974; Tankersley 1984; Turney and Blume

3.8

3.8

0.5
0.5
0.8

2.3
0.2
1.7
1.2
0.5
0.5
.

2004; Poole et al. 2010). The earlier onset and decline

in mineral lick visitation by the Resident in this study
may be due to the fact that he was a resident, that he
was the only adult male in my study, or to particular
1
5
2
8
5
7
12
11
4
3
13
4
35
45
100
n

characteristics of the licks and associated terrain.

Notably, other reports have documented earlier lick
31.8
51.5
37.1
51.5
7.9
5.4
7.9
2.0
1.7
7.5
7.6
4.0
7.6
2.9
51.5
Max

use for males: for Mountain Goats (Hebert and Cowan

At Lick

1971; Poole et al. 2010; J. Mainguy, personal com-

munication, 2008); Moose (Alces alces), Fraser and
Min
31.8
0.5
33.0
0.5
0.1
2.0
0.1
0.1
0.4
0.2
0.6
1.3
0.1
0.1
0.1

Hristienko 1981; Tankersley and Gasaway 1983; and

Dall Sheep (Ovis dalli dalli), Tankersley 1984. In
Median1

mountain environments, snow may impede and hence

2.4b

1.3b
0.8b
32.4a
31.8
18.3
35.0

1.5
2.4

0.4
1.1
2.5
1.5
1.6

1.0

delay long-distance movements compared to lower ele-

vations and may explain the later peak in mineral lick
visitation by Migrant Mountain Goats.
indicate statistically significant differences between Types.

Although I did not find significant differences

Goats in Washington, 13 September 2003–19 September 2007.

among Types for Durations of States other than At

0
2
1
3
5
2
7
10
3
1
11
3
28
26
64
n

Lick, it seems likely that this was due to low statistical

power (given the small sample sizes) rather than the
Max

14.3
2.2
14.3
4.0
1.8
4.0
2.4
4.1
1.0
2.3
1.5
4.1
3.0
14.3

lack of actual differences. For States of Toward and

.
Toward

Away, Sojourners had Durations approximately half

those of Migrants, Commuters had Durations approx-
Min

4.1
2.2
2.2
0.3
1.6
0.3
0.1
1.3
1.0
0.4
0.8
0.1
0.1
0.1

imately half those of Sojourners, and the Resident

.

had Durations approximately half those of Commuters

(Table 2). A similar progression was evident in the non-
Median

9.2
2.2
4.1
1.0
1.7
1.6
0.3
1.5
1.0
0.9
1.1
1.0
0.5
1.0

significant Distances, except that Sojourner and Com-

.

muter Distances were approximately equal (Table 4).

Mineral lick use can be considered in a cost-benefit
framework, in which the benefit is the chemical con-
038GMM
051GPM

045MRF
024KRF
037HRF
033GPF
034GPF

052GPF
053GPF
054GPF
055GPF

stituents available from the lick (Kreulen 1985; Klaus

Goat

and Schmid 1998) and the costs are the energetic costs
All

All

All

All

of traveling Toward and Away, reduction of foraging

opportunity, and increased predation risk associated
Commuter
Sojourner

with travel and remaining in the vicinity of the lick.

Resident
Migrant

1letters

Foraging opportunity was probably reduced in the

Type

All

vicinity of most licks. The Deadhorse Point and French

2010 RICE: LICK VISITATION BY MOUNTAIN GOATS 233

FIGURE 4. Weekly means of the daily proportion of Mountain Goats At Lick for each Type in Washington, 13 September
2003–19 September 2007.

TABLE 3. Median, minimum, and maximum Intervals (difference, in days, between the start of Toward and the end of the
previous Away) between mineral lick visits by Mountain Goats taking place in the same year and between years (over winter)
in Washington, 13 September 2003–19 September 2007.
Same year Between years
Type Goat Median Min Max n Median Min Max n
Migrant 033GPF . . . 0 . . . 0
034GPF . . . 0 316.7 312.4 321.0 2
051GPM . . . 0 317.8 317.8 317.8 1
All . . . 0 317.8 312.4 321.0 3
Sojourner 024KRF 64.8 51.0 78.5 2 237.4 224.5 250.2 2
037HRF 16.4 16.4 16.4 1 318.9 318.9 318.9 1
All 51.0a 16.4 78.5 3 250.2 224.5 318.9 3
Commuter 045MRF 5.5 0.9 89.6 9 . . . 0
052GPF 15.5 6.3 16.9 3 . . . 0
053GPF 12.9 12.9 12.9 1 . . . 0
054GPF 6.0 1.5 16.2 10 290.5 290.5 290.5 1
055GPF 5.2 1.7 16.5 3 . . . 0
All 6.3b 0.9 89.6 26 290.5 290.5 290.5 1
Resident 038GMM 3.1b 0.4 93.5 23 143.1 124.5 161.6 2
All All 5.6 0.4 93.5 52 290.5 124.5 321.0 9
1letters indicate statistically significant differences between Types

Cabin Creek licks were in timbered areas which may risk would be considered higher for visits to Deadhorse
have less forage available than open areas, although Point and French Cabin Creek licks, but less for the
this was not the case with the subalpine licks (Gardner subalpine licks. Also, Gamma Ridge differed from the
Mountain, Gamma Ridge, and Mineral Mountain). other lick areas in that extensive alpine meadows and
Also, it is highly likely that forage in the vicinity of escape terrain occurred in the vicinity of the licks. How-
licks is over-utilized by high concentrations of visiting ever, Migrants visiting the Gamma Ridge licks traveled
Mountain Goats (Cowan and Brink 1949). For Moun- though timbered terrain on route to the subalpine licks
tain Goats, predation risk probably increases in unfa- there. The artificially high concentrations of Mountain
miliar terrain and with restricted visibility in forested Goats in the vicinity of licks may also increase preda-
habitats (Cowan and Brink 1949; Festa-Bianchet et al. tion risk. The tradeoffs in these factors have apparently
1994; Côté and Beaudoin 1997). As such, predation resulted in the lick visit Types, where the Mountain
 234
TABLE 4. Median, maximum, and minimum Distance (km) and movement rate (km/hr) for individual mountain goats moving Toward and Away from a mineral lick from the first fix of THE CANADIAN FIELD-NATURALIST Vol. 124

TABLE 5. Rate of movement (m/hr) by State and Type during

Away the lick season (01Jun-15Aug) for mountain goats visiting
1
3
1
5
5
3
8

71
10
4
2
12
4
32
26
mineral licks.
Mean1
n

Effect Level 95%CI n

Toward
0
2
1
3
5
2
7

65
10
3
1
11
4
29
26
State Toward 70.9a 54.9–91.5 547
Away 52.0b 40.6–66.5 787
Excursion 34.1bcd 23.5–49.4 95
At Lick 27.1c 21.5–34.3 1864
0.20
0.31
0.32
0.32
0.16
0.28
0.28

0.73
0.73
0.20
0.30
0.62
0.25
0.73
0.50
Max

None 23.7d 18.9–29.8 5131

Type Commuter 44.1 33.5–58.1 3655
Migrant 37.8 27.4–52.2 2097
0.20
0.11
0.32
0.11
0.10
0.15
0.10

0.01
0.13
0.08
0.14
0.10
0.14
0.08
0.01
Min
Away

Resident 36.9 22.2–61.6 1179

Sojourner 34.3 23.4–50.5 1493
Median

1
letters indicate statistically significant differences
0.20
0.13
0.32
0.20
0.13
0.16
0.14

0.16
0.19
0.12
0.22
0.23
0.18
0.19
0.10
Movement Rate

Goat may: (1) visit the lick infrequently and remain in

the vicinity for an extended period because the costs
of travel are high and habitat in the vicinity of the lick
Max

0.18
0.29
0.29
0.92
0.34
0.92

1.45
1.45
0.34
0.67
0.86
0.40
1.45
0.75

is acceptable (Migrant); (2) visit the lick infrequently

.

and remain in the vicinity for a short period because

the costs of travel are high and habitat in the vicinity
Toward

0.05
0.29
0.05
0.06
0.33
0.06

0.04
0.22
0.10
0.67
0.26
0.21
0.10
0.04
Min

of the lick is unacceptable (Sojourner); or (3) visit the

.

lick frequently and remain in the vicinity for a short

period because the costs of travel are low (Commuter
Median

0.12
0.29
0.18
0.22
0.34
0.33

0.30
0.61
0.23
0.67
0.40
0.26
0.40
0.23

and Resident).
.

Given the low number of Mountain Goats and licks

in this study, it is difficult to be certain whether these
Types are artificial divisions along a continuum of res-
ponses or natural categories emerging from the trade-
29.4
18.2
7.9
29.4
6.7
12.9
12.9

29.4
6.5
12.1
18.2
12.9
7.6
18.2
5.1
Max

offs discussed above. However, other accounts of min-

eral lick visitation suggest they can be fit into these
Types. Hebert and Cowan (1971) indicated Mountain
Away

29.4
15.9
7.9
7.9
5.3
12.1
5.3

0.6
4.2
7.7
16.1
6.7
4.8
4.2
0.6
Min
the movement to the last in Washington, 13 September 2003–19 September 2007.

Goats visited licks briefly once a year, which would

be Sojourners. Singer and Doherty (1985) described
frequent visits by Mountain Goats from Glacier
Median
29.4
17.9
7.9
17.9
6.4
12.5
6.7

5.3
4.7
8.7
17.1
7.9
7.4
7.5
2.0

National park (Commuters) but suspected that Moun-

tain Goats coming from Flathead National Forest vis-
Distance

ited only once per year for < 2 weeks (Sojourners). The
Bighorn Sheep (Ovis canadensis) studied by Hnilicka
et al. (2002) made fortnightly visits to the lick through-
Max

18.3
15.6
18.3
8.9
14.0
14.0

18.3
12.8
10.3
16.8
15.1
8.1
16.8
6.9

out the summer (Commuters), whereas Dall Sheep

.

(Ovis dalli) in Alaska visited the licks primarily during

the transition from winter to summer range (Heimer
Toward
Min

17.6
15.6
15.6
5.5
13.4
5.5

1.5
4.1
8.1
16.8
7.3
7.3
4.1
1.5

1974; Sojourners). Adult Moose (Alces alces) studied

.

by Fraser and Hristienko (1981) were evidently Sojour-

ners and Commuters, while young male Moose were
Median

18.0
15.6
17.6
5.5
13.7
5.6

5.5
4.6
10.1
16.8
8.0
7.7
7.8
2.3

Migrants. White-tailed Deer (Odocoileus virginianus)

.

studied by Wiles and Weeks (1986) had licks within

their usual ranges (Residents) or traveled frequently to
nearby licks (Commuters).
038GMM

Comparisons of licks soils across studies can be dif-

051GPM

045MRF
024KRF
037HRF
033GPF
034GPF

052GPF
053GPF
054GPF
055GPF

ficult due to inconsistent choices of which constituents

Goat

All

All

All
All

to measure and differing methods of measurement

(Klaus and Schmid 1998). Some of my measurements
can be compared with those reported by Jones and
Commuter
Sojourner

Resident

Hanson (1985) for the geometric mean of 18 mineral

Migrant

licks used by Mountain Sheep and Mountain Goats:

Type

All

Na = 1.67; Ca = 27.85; Mg = 5.69. The significantly

2010 RICE: LICK VISITATION BY MOUNTAIN GOATS 235

TABLE 6. Mean concentrations of Calcium (Ca), Potassium (K), Magnesium (Mg), Sodium (Na), and sulphate (SO4) in mineral
lick (Site) and reference (Ref) soil samples in Washington and t–test evaluations of differences for each lick. Means are
back transformed estimates from the average of log–transformed values. Effect is the difference in the averages of the
log–transformed values (site – reference) and the t–test evaluates the hypothesis of site > reference. meq = milliequivalents.
Mineral Concentration Samples
Chemical Lick Site Ref Site Ref Effect t P
Ca GAM1 7.49 10.10 2 4 –0.300 0.706 0.757
(meq/100g) GAM2 6.77 5.96 3 3 0.127 –0.317 0.377
GAM3 5.70 6.60 1 4 –0.146 0.267 0.604
GAM5 6.80 4.83 1 4 0.343 –0.627 0.268
GAR1 3.12 2.49 2 8 0.224 –0.579 0.284
GAR2 4.71 2.82 2 8 0.512 –1.322 0.098
K GAM1 245.73 138.05 2 4 0.577 –1.927 0.032*
(ppm) GAM2 191.00 133.47 3 3 0.358 –1.271 0.107
GAM3 317.00 127.17 1 4 0.913 –2.365 0.012*
GAM5 90.00 75.40 1 4 0.177 –0.458 0.325
GAR1 97.86 55.35 2 8 0.570 –2.086 0.023*
GAR2 47.29 53.93 2 8 –0.131 0.481 0.683
Mg GAM1 4.05 5.71 2 4 –0.343 1.037 0.846
(meq/100g) GAM2 4.79 4.76 3 3 0.007 –0.022 0.491
GAM3 4.60 3.54 1 4 0.261 –0.612 0.273
GAM5 1.10 1.01 1 4 0.082 –0.193 0.424
GAR1 0.39 0.22 2 8 0.582 –1.930 0.032*
GAR2 0.30 0.24 2 8 0.203 –0.672 0.253
Na GAM1 2.39 0.12 2 4 3.003 –11.267 <0.001*
(meq/100g) GAM2 0.87 0.15 3 3 1.764 –7.020 <0.001*
GAM3 4.75 0.17 1 4 3.305 –9.605 <0.001*
GAM5 0.54 0.22 1 4 0.888 –2.580 0.008*
GAR1 0.22 0.12 2 8 0.621 –2.554 0.008*
GAR2 0.26 0.17 2 8 0.427 –1.755 0.045*
SO4 GAM1 208.94 35.87 2 4 1.762 –2.205 0.018*
(ppm) GAM2 98.11 29.14 3 3 1.214 –1.611 0.059
GAM3 143.13 27.32 1 4 1.656 –1.605 0.059
GAM5 32.13 20.87 1 4 0.431 –0.418 0.339
GAR1 5.13 4.71 2 8 0.085 –0.116 0.454
GAR2 2.58 1.31 2 8 0.674 –0.923 0.182
*Significant at α = 0.05

higher concentrations of Na at licks and higher concen- It is noteworthy that all seven Mountain Goats vis-
trations, but less extreme differences for other chemi- iting Gamma Ridge (Migrants and Sojourners) crossed
cals are similar to the results of other studies (e.g. the crest of the Cascade Range during the movements
Weeks and Kirkpatrick 1976; Tankersley 1984; Klein Toward and Away from the licks (Figure 2). While
and Thing 1989; Tracy and McNaughton 1995; Klaus doing so, they also crossed from one national forest
and Schmid 1998; Mincher et al. 2008). This supports to another, from one Department of Fish and Wildlife
the conclusion that Na is the main reason the mountain administrative region to another, and from one county
goats in this study visited mineral licks. My results do of Washington to another. 037HRF also crossed region
not support the hypothesis that Mg is a chemical sought and county boundaries during mineral lick visits. Con-
after at licks (Jones and Hanson 1985; Heimer 1988; sequently, coordination among administrative units is
Klaus and Schmid 1998). The preponderance of visits a necessary part of managing these Mountain Goats
to mineral licks in the late spring and early summer and the mineral licks they use. Nevertheless, little is
suggests that the detoxification/acidosis function of known about the degree to which disturbances (log-
Na (Foley et al. 1995) is not the driver for mineral ging, recreation, road construction, trail development)
lick visitation because secondary compounds are more may impact mineral lick movements and this may vary
prevalent in browse than grasses and forbs (Festa- among Mountain Goat populations. Poole et al. (2010)
Bianchet 1988), and Mountain Goat diet is typically gave anecdotal accounts of logging modifying, but not
mostly forbs and graminoids in the summer with the inhibiting Mountain Goat lick visitation, but at a road
most browse consumed in the winter (Fox et al. 1989). inhibited movements for about a year. It would seem
236 THE CANADIAN FIELD-NATURALIST Vol. 124

prudent to limit logging operations and road building Alaska. General Technical Report PNW-GTR-246, U.S.
along known mineral lick travel routes to avoid times Department of Agriculture and Forest Service, Pacific
of high lick visitation. Northwest Research Station, Portland, Oregon, USA.
Fraser, D., and H. Hristienko. 1981. Activity of moose and
white-tailed deer at mineral springs. Canadian Journal of
Acknowledgments Zoology 59: 1991–2000.
This study was a cooperative one and benefited from Hebert, D., and I. McT. Cowan. 1971. Natural salt licks as
the contributions of the U.S. Forest Service (D. Gay, part of the ecology of the mountain goat. Canadian Jour-
P. Reed), Western Washington University (T. Beus nal of Zoology 49: 605–610.
and A. Shirk), the Sauk-Suiattle Tribe (C. Danilson and Heimer, W. E. 1974. The importance of mineral licks to Dall
E. Peterson), the National Park Service (J. Schaberl) sheep in interior Alaska and its significance to sheep man-
and other employees of the Department of Fish and agement. Biennial Symposium of the Northern Wild Sheep
Wildlife (G. Falxa, S. Fitkin, B. Hall, R. Milner, and and Goat Council 3: 49–64.
Heimer, W. E. 1988. A magnesium-driven hypothesis of Dall
W. Moore). T. Beus, B. Kertson, L. Leschner, J. Main-
sheep mineral lick use: preliminary tests and management
guy, K. Poole, S. Pearson, E. Rice, K. White, and G. relevance. Biennial Symposium of the Northern Wild Sheep
Wiles provided comments on the manuscript. Support and Goat Council 6: 269–279.
for this work was from U.S. Fish and Wildlife Serv- Hnilicka, P., J. Mionczynski, B. Mincher, J. States, M.
ice under the State Wildlife Grants Program and Fed- Hinschberger, S. Oberlie, C. Thompson, and B. Yates.
eral Aid in Wildlife Restoration, the Wildlife Research 2002. Bighorn sheep lamb survival, trace minerals, rainfall
Program of Seattle City Light, the Sauk-Suiattle Tribe, and air pollution: Are there any connections? Northern
and the United States Forest Service Challenge Cost Wild Sheep and Goat Council 13: 69–94.
Jones, R. L., and H. C. Hanson. 1985. Mineral licks, geo-
Share program.
phagy, and biochemistry of North American ungulates.
Iowa State University Press, Ames.
Literature Cited Kennedy, J. F., J. A. Jenks, R. L. Jones, and K. J. Jenkins.
Atwood, T. C., and H. P. Weeks, Jr. 2002. Sex- and age- 1995. Characteristics of mineral licks used by white-tailed
specific patterns of mineral lick use by white-tailed deer deer (Odocoileus virginianus). American Midland Natu-
(Odocoileus virginianus). American Midland Naturalist ralist 134: 324–331.
148: 289–296. Klaus, G., and B. Schmid. 1998. Geophagy at natural licks
Ayotte, J. B., K. L. Parker, J .M. Arocena, and M. P. and mammal ecology: A review. Mammalia 62: 481–497.
Gillingham. 2006. Chemical composition of lick soils: Klein, D. R., and H. Thing. 1989. Chemical elements in
functions of soil ingestion by four ungulate species. Journal mineral licks and associated muskoxen feces in Jameson
of Mammalogy 87: 878–888. Land, northeast Greenland. Canadian Journal of Zoology
Comer, P., D. Faber-Langendoen, R. Evan, S. Gawler, C. 67: 1092–1095.
Josse, G. Kittel, S. Menard., M. Pyne, M. Reid, K. Kreulen, D. A. 1985. Lick use by large herbivores: a review of
Schulz, K. Snow, and J. Teague. 2003. Ecological systems benefits and banes of soil consumption. Mammal Review
of the United States: a working classification of U.S. ter- 15: 107–123.
restrial systems. NatureServe, Arlington, Virginia. Lewis, J. S., J. L. Rachlow, E. O. Garton, and L. Vierling.
Côté, S. D., and C. Beaudoin. 1997. Grizzly bear (Ursus arc- 2007. Effects of habitat on GPS collar performance: using
tos) attacks and nanny-kid separation on mountain goats data screening to reduce location error. Journal of Applied
(Oreamnos americanus). Mammalia 61: 614–617. Ecology 44: 663-671.
Cowan, I. McT., and V. C. Brink. 1949. Natural game licks Mincher, B. J., R. D. Ball, T. P. Houghton, J. Mionczynski,
in the Rocky Mountain national parks of Canada. Journal and P. A. Hnilicka. 2008. Some aspects of geophagia in
of Mammalogy 30: 279–397. Wyoming bighorn sheep (Ovis canadensis). Wildlife
Dormaar, J. F., and B. D. Walker. 1996. Elemental content Research 54: 193–198.
of animal licks along the eastern slopes of the Rocky Moe, S. R. 1993. Mineral content and wildlife use of soil
Mountains in southern Alberta, Canada. Canadian Journal licks in southwestern Nepal. Canadian Journal of Zoology
of Soil Science 76: 509–512. 71: 933–936.
Festa-Bianchet, M. 1988. Birthdate and survival in bighorn Poole, K. G., K. D. Bachmann, and I. E. Teske. 2010. Min-
lambs (Ovis canadensis). Journal of Zoology 214: 653- eral lick use by GPS radio-collared mountain goats in
661. southeastern British Columbia. Western North American
Festa-Bianchet, M., and S. D. Côté. 2008. Mountain Goats: Naturalist 70:208-217.
Ecology, Behavior, and Conservation of an Alpine Ungu- Poole, K. G., and D. C. Heard. 2003. Seasonal habitat use
late. Island Press. Washington D.C., USA. and movements of mountain goats, Oreamnos americanus,
Festa-Bianchet, M., M. Urquhart, and K. G. Smith. 1994. in east-central British Columbia. Canadian Field-Naturalist
Mountain goat recruitment: kid production and survival 117: 565–576.
to breeding age. Canadian Journal of Zoology 72: 22–27. Rice, C. G. 2008. Seasonal altitudinal movements of moun-
Foley, W. J., S. McLean, and S. J. Cork. 1995. Conse- tain goats. Journal of Wildlife Management 72: 1706–
quences of biotransformation of plant secondary metabo- 1716.
lites on acid-base metabolism in mammals—a final com- Rice, C. G., and B. Hall. 2007. Hematologic and biochemi-
mon pathway? Journal of Chemical Ecology 21: 721-743. cal reference intervals for mountain goats (Oreamnos amer-
Fox, J. L., C. A. Smith, and J. W. Schoen. 1989. Relation icanus): effects of capture conditions. Northwest Science
between mountain goats and their habitat in southeastern 81: 206–214.
2010 RICE: LICK VISITATION BY MOUNTAIN GOATS 237

Rice, C. G., K. J. Jenkins, and W. Chang. 2009. A sighta- central British Columbia. Biennial Symposium of the
bility model for mountain goats. Journal of Wildlife Northern Wild Sheep and Goat Council 14: 64.
Management 73: 468–478. Weeks, H. P., Jr., and C. M. Kirkpatrick. 1976. Adaptations
SAS Institute. 2007. JMP User Guide, Release 7. SAS Insti- of white-tailed deer to naturally occurring sodium deficien-
tute, Cary, North Carolina, USA. cies. Journal of Wildlife Management 40: 610–625.
Singer, F. J., and J. L. Doherty. 1985. Movements and Wells, A. 2006. Global Positioning System (GPS) Bias Cor-
habitat use in an unhunted population of mountain goats, rection and Habitat Analysis of Mountain Goats Oreamnos
Oreamnos americanus. Canadian Field-Naturalist 99: 205– americanus in the Cascades of Washington State, USA M.
217. Sc. Thesis, Western Washington University, Bellingham.
Tankersley, N. G. 1984. Mineral lick use by Dall sheep in Wiles, G. J., and H. P. Weeks, Jr. 1986. Movements and use
the Watana Creek Hills, Alaska. Biennial Symposium of patterns of white-tailed deer visiting natural licks. Journal
the Northern Wild Sheep and Goat Council 4: 211–230. of Wildlife Management 50: 487–496.
Tankersley, N. G., and W. C. Gasaway. 1983. Mineral lick Wright, W. 1977. Ecology of the Cascade Mountain Goat,
use by moose in Alaska. Canadian Journal of Zoology 61: Mount Baker-Snoqualmie National Forest, Washington.
2242–2249. M.Sc. Thesis, Western Washington State University, Bel-
Tracy, B. F., and S. J. McNaughton. 1995. Elemental analy- lingham.
sis of mineral lick soils from the Serengeti National Park, Zar, J. H. 1996. Biostatistical Analysis. 3rd edition. Prentice
the Konza Prairie and Yellowstone National Park. Ecog- Hall, Upper Saddle River.
raphy 18: 91–94.
Turney, L., and R. Blume. 2004. Daily and seasonal move- Received 24 June 2010
ment patterns of mountain goats to a mineral lick in north- Accepted 27 September 2010