USGS Home
Contact USGS
Search USGS

We used Jolly-Seber models with time-specific survival and capture probabilities to estimate numbers of females (Model A of Pollock et al. 1990) because they fit better than models with constant survival probabilities (Model B: all females χ²₅ = 12.3, P = 0.03; adults only χ²₃ = 10.9, P = 0.01; truncated capture histories χ²₃ = 8.6, P = 0.03) and models with constant survival and recapture probabilities (Model D: all females χ²₁₃ = 27.3, P = 0.01; adults only χ²₁₃ = 25.84, P = 0.02; truncated capture histories χ²₈ = 14.4, P = 0.07). Further, goodness-of-fit was satisfactory only for Model A (all females χ²₉ = 4.07, P = 0.91; adults only χ²₅ = 3.52, P = 0.62; truncated capture histories χ²₄ = 3.28, P = 0.51). Estimated numbers of females (Table 1) were relatively stable from 1968 to 1970, increased abruptly during 1971-72, remained high through 1974, then declined abruptly (Fig. 2).

Table 1.  Estimated numbers of black bears using a 218-km² area near Cold Lake Alberta, 1968-77. Estimates of total numbers (all bears) are from Young and Ruff (1982) unless otherwise noted. Estimates for females are from this study and are based on Jolly-Seber models with time-specific survival and recapture probabilities.
Year	All bears	All females		Adult females
			SE		SE
1968	84	— a	— a	— a	— a
1969	71	15.9	2.88	16.1	3.11
1970	92	21.4	4.36	17.1	3.51
1971	75	31.6	9.49	18.9	5.80
1972	117	45.8	11.38	40.4	15.63
1973	175	46.7	15.60	31.5	19.42
1974	136	59.1	24.61	41.4	27.02
1975	137	25.5	5.85	22.9	7.24
1976	123 b	34.2	5.26	25.0	— c
1977	99 b	— c	— c	— c	— c
a Not estimable by Jolly-Seber method b R.L. Ruff, unpublished data c Limited data precluded estimation

Fig. 2.  Trends in estimated numbers of bears using the CLSA during 1968-77, standardized for comparison. Estimates from Young and Ruff (1982) are based on male and female bears aged 2 years (). Others are Jolly-Seber estimates for females (age 4 years []; age 2 years []; captures prior to 1975, age 2 years []).

The abrupt nature of the increase in estimated numbers of female bears suggests a change in spatial distribution, rather than a change in reproduction or survival. Survival records of radiomarked bears suggest that mortality may have been partially responsible for the abrupt decline in 1975 (Table 2). K-M estimates of mortality caused by humans were similar for radiomarked male and female bears (M:F risk ratio = 1.05, χ²₁ = 0.014, P = 0.91) and were not significantly lower for subadults than for adults (ad:sa risk ratio = 0.80, χ²₁ = 0.20, P = 0.66).

Human activity was the proximate cause of at least 25, and probably 28, of 30 deaths that befell radiomarked bears during 1974-77. Principal causes of mortality were depredation control (11 deaths) and a combination of legal hunting by native people and poaching (11 deaths), which could not be distinguished. Legal sport hunting, a drug overdose, a collision with a train, another bear, and other natural causes each caused 1 death. Fates of 3 bears could not be confirmed by recovery of a carcass or radiocollar, but each was rumored to have been shot.

The number of subadult males captured on the CLSA was larger during each year of 1972-75 (range = 21-25) than during any year of 1968-71 or 1976-77 (range = 2-17; Table 3). We believe this result indicates a difference in the number of subadults that were at least temporarily present on the study area because estimated capture probabilities of subadults were not significantly higher (χ²₁ = 1.13, P = 0.29) during the post-removal period (_{post, sa} = 0.55, SE = 0.11) than during the pre-removal period (_{pre, sa} = 0.72, SE = 0.13).

Moreover, estimated settling rates did not differ enough between periods (_{pre, sa} = 0.49, SE = 0.10; _{post, sa} = 0.49, SE = 0.07) to offset the increase in captures. Thus, numbers of subadult males using the study area were probably larger during 1972-75 than during other years.