Swift fox (Vulpes velox) COSEWIC assessment and status report 2009: chapter 7


Population Sizes and Trends

Search Effort

Population censuses using live–trapping and sign surveys have been conducted every five years between 1996 and 2006 during winter months (November–February). The area surveyed was determined by the Canadian Swift Fox Recovery Team based on habitat criteria (Cotterill 1997a). The survey area was expanded in subsequent surveys to track range expansion. Surveys enabled population estimates for 1996–1997, 2000–2001, and 2005–2006, although confidence limits around these estimates are not yet available.

The study area for the 2005–2006 census extended roughly from Manyberries, AB, east to Glentworth, SK (106° 33’ W − 110° 50’ W), and from Eastend, SK, south to Dodson, MT (48° 24’ N − 49° 31’ N). In total, 191 townships, covering 12 717 km² were surveyed (Moehrenschlager and Moehrenschlager 2006). This included 4061 km² in 62 townships for the Border population, 2550 km² in 38 townships in the GNP population, and 6106 km² in 91 townships in Montana. The area surveyed was 26% larger than in 2000–2001.

Due to logistical constraints associated with winter trapping, traditional population estimation methods are not possible from census data, thus catch–and–release live–trapping was employed, based on a calibration–based census design (see Cotterill 1997a). In each township, one 5 km transect along a continuous trail was laid out near the township centre. Traps were placed along fences or on hilltops at approximately 1 km intervals on the transect (Moehrenschlager and Moehrenschlager 2006). Home range data were used to estimate the area sampled per trap, as well as trappability, data necessary for estimating abundance. A correction factor was used to account for trapping success and trappability of animals for which a home range estimate was available (correction factor = [foxes available for capture] / [captured foxes from known home ranges = 3.5]. A fox was considered available for capture when more than 50% of its home range fell within one home range radius of a trap). The correction factor was applied to all population calculations (Cotterill 1997a).

Each transect received 18 trap nights (i.e., six traps on three nights), amounting to a total effort of 3438 trap nights (Moehrenschlager and Moehrenschlager 2006). This was a 28% increase over the 2000–2001 census and a tripling of effort from 1996–1997. Incidental sightings of Swift Foxes were also recorded; however, unlike in the previous census, spotlighting and snow–tracking were not used.

Abundance

Based on live–trapping data from 2005–2006, there are an estimated 513 Swift Foxes in the Border population and 134 in GNP (Moehrenschlager and Moehrenschlager 2006). The neighbouring northern Montana population had an estimated 515 foxes. The census data have not yet undergone bootstrap analyses, and thus, confidence intervals are not available. Thus population estimates from 2000–2001 and 2005–2006 are overly precise and can only be considered as the best fit to the available data. In total, the Canadian population is estimated to be 647. Because surveys were conducted just before the onset of the breeding season and both males and females reach sexual maturity in one year, this count refers to number of mature individuals.

Fluctuations and Trends

Based on the 1991–1992 survey, the Canadian population was estimated to be 225 individuals (range: 150–300; Brechtel et al. 1993). This increased to 289 in 1996–1997 (95% confidence interval: 179–412; Cotterill 1997b). By 2000–2001 the population had more than doubled to 656 (Moehrenschlager and Moehrenschlager 2001) and the Border and GNP populations had become contiguous through northern Montana. With an estimated 221 Swift Foxes in the Montana border population, the overall Alberta–Saskatchewan–Montana population was approximately 887. The 2005–2006 census suggests the Canadian population has remained stable while the Montana population increased to 515 foxes, for a combined 2006 population estimate of 1162 Swift Foxes (Moehrenschlager and Moehrenschlager 2006).

Since 2001, the Border population has expanded from 18 to 32 townships, a 78% increase (Figure 7), although the distribution has declined in the past five years (from 38 to 32 townships). However, when the data from the 48 townships included in both surveys are used, there has been a slight increase in the proportion of townships where Swift Foxes were detected, from 58 to 60% (Moehrenschlager and Moehrenschlager 2006).

The distribution of the GNP population increased to 43% (13 of 30) of townships surveyed in 2000–2001. The 2005–2006 census showed a 43% contraction in the GNP distribution, with foxes detected in only 21% (8 of 38) of townships (Figure 7). Although the distribution in GNP in 2005–2006 was less than half from five years before, for the 30 townships included in both surveys, the decrease was only 10% (Moehrenschlager and Moehrenschlager 2006).

Although the census estimates indicate increasing populations, they are confounded by the survey design, due to the inclusion of previously unsurveyed areas. Thus, the apparent increases are likely in part due to greater trapping effort.

In 2000–2001 the density of Swift Foxes in the Border and GNP populations was 6.7 per 100 km². Including the northern Montana population, the density was estimated at 5.1 per 100 km² (Moehrenschlager and Moehrenschlager 2001). Population densities were similar in 2005–2006, with the Canadian population at 5.5 per 100 km² and with Montana included 5.4 per 100 km² (Moehrenschlager and Moehrenschlager 2006).

In the latest census it was clear that improvements to the status of Swift Fox population at large were mostly attributable to the Montana portion. Specifically, over half of Montana townships that had not had captures in 2001/2 did have captures in 2005/2006, whereas this was the case for only 19.5% of Canadian townships. Similarly, while the population estimate for Canadian animals more than doubled between the first two survey periods, they did not change thereafter. By contrast, population estimates on the Montana side of the border have steadily increased in each census (Cotterill 1997a; Moehrenschlager and Moehrenschlager 2001; 2006).

Age structure and sex ratios provide insight into survival and productivity and are indicators of reintroduction success. In Canada, the proportion of adults captured increased from 50% in 1996–1997 (n=32; Cotterill 1997b), to 60% in 2000–2001 (n=107; Moehrenschlager and Moehrenschlager 2001). Age structure based on 2005–2006 census data have not been analyzed. Sex ratios shifted from 63% males in 1996–1997 (n=32; Cotterill 1997b), to 43% in 2000–2001 (n=108; Moehrenschlager and Moehrenschlager 2001), and was close to parity in 2005–2006 (52% males; n=195; Moehrenschlager and Moehrenschlager 2006). Sex ratios were not significantly different in the Border, GNP, and northern Montana populations (Moehrenschlager and Moehrenschlager 2006).

Border Population

The Border population was established by reintroductions into the Lost River Ranch area on the Alberta–Saskatchewan border between 1983 and 1997. A 1989–1991 survey yielded a population estimate of 150–250 foxes (Brechtel et al. 1993). The population decreased to 100–135 by 1994 (Mamo 1994), but increased to 192 (95% CI: 93–346) by 1996–1997 (Cotterill 1997a). The 1999 survey showed the population had increased by 38% since 1997 (Moehrenschlager and Moehrenschlager 1999). By 2000–2001, the population was 560 (Moehrenschlager and Moehrenschlager 2001).

In 2000–2001, Swift Fox density in the Border population was relatively high at 9.3 foxes/100 km² (Moehrenschlager and Moehrenschlager 2001). It decreased to 7.2 in 2005–2006. The most plausible explanation for the decline is dispersal by foxes into Montana, as this population increased by 77% during the period (Moehrenschlager and Moehrenschlager 2001, 2006).

The Border population had similar proportions of adults in 1996–1997 (54%; n=24; Cotterill 1997a) and 2000–2001 (59%; n=94; Moehrenschlager and Moehrenschlager 2001). The proportion of males decreased from 54% in 1996–1997 (n=24; Cotterill 1997a) to 41% in 2000–2001 (n=95; Moehrenschlager and Moehrenschlager 2001), a change that is likely due to demographic stochasticity.

GNP Population

The GNP population was established by reintroductions into the East and West Blocks of the park and the Wood Mountain area between 1990 and 1997. In 1991, the GNP population had less than 50 individuals (Brechtel et al. 1993). Two years later, the population was estimated to be 25 (range: 15–50; Hjertaas 1994). The 1996–1997 census indicated a population of 87 (Cotterill 1997a). The population increased to 96 by 2000–2001 (Moehrenschlager and Moehrenschlager 2001) and then to 134 by 2005–2006 (Moehrenschlager and Moehrenschlager 2006), an increase of 54% over 10 years.

Swift Fox densities in the GNP population were similar between 2000–2001 and 2005–2006 (2.4 and 2.7 per 100 km²; respectively; Moehrenschlager and Moehrenschlager 2001, 2006).

The ratio of adults to juveniles in GNP increased from 1996–1997 (38%; n=8; Cotterill 1997a) to 2000–2001 (69%; n=13; Moehrenschlager and Moehrenschlager 2001). The percentage of males declined from 88% in 1996–1997 (n=8; Cotterill 1997a) to near parity in 2000–2001 (54%; n=13; Moehrenschlager and Moehrenschlager 2001). Age structure and sex ratios for the 2005–2006 data have not been analysed.

Rescue Effect

With 100% of trapped individuals wild–born in the 2005–6 survey, it is clear that reintroduction programs in Alberta and Saskatchewan have produced self–sustaining populations. Moreover, this effort has resulted in the establishment of Swift Foxes in northern Montana, which appear to be increasing in numbers and dispersing back into Canada. Due to the connection between the Border and the GNP population through north–central Montana, the potential for a limited rescue effect therefore exists. Continued improvements in the status of Montana Swift Fox may already be contributing to the stable status of this species in Canada, where habitat degradation through energy development and associated roads is ongoing. There is little or no potential for rescue beyond the reintroduced populations directly across the US border, because the contiguous Alberta–Saskatchewan–Montana populations are disjunct from core Swift Fox range in the US.


Figure 7: Comparison of Townships with Swift Fox Captures Between 2000–2001 and 2005–2006

Map indicating locations of townships with Swift Fox captures, lack of captures, and/or sightings during the 2000–01 and 2005–06 censuses in Alberta, Saskatchewan and Montana.

From Moehrenschlager and Moehrenschlager 2006.

Page details

Date modified: