Western harvest mouse (Reithrodontomys megalotis) COSEWIC assessment and update report: chapter 7

• Previous page
• Table of contents
• Next page

Population Sizes and Trends

Search effort

Trapping has been the primary method used to determine the presence/absence and abundance estimates for this species in Canada. Search effort is most often described in terms of trap-nights. Kill trapping along transects was commonly used to determine presence/absence (e.g., Reynolds et al. 1999), whereas 1 ha live-trapping grids were used to estimate abundance (e.g., Sullivan and Sullivan 2004, W. Klenner unpubl. data). Nagorsen (1995) used a combination of live- and kill- trapping along 500 m transects to assess presence/absence throughout BC.

Abundance

Even at a localized scale (e.g., 1 ha trapping grids), there are few estimates of western harvest mouse abundance. Although densities as high as 60 animals/ha have been reported in grasslands in the US (Whitford 1976), the mouse is naturally rare within Canada, typically comprising <10% of the total number of small mammals within a given habitat (Nagorsen 1995, Cannings et al. 1999, Reynolds et al. 1999). There are no provincial or national estimates of abundance.

British Columbia

At Prairie Valley (near Summerland), mean density of western harvest mice over fall-winter months ranged from 4.7/ha in poor-quality habitat, to 54.3/ha in high-quality habitat (Sullivan and Sullivan 2006a; Table 1). During December 1994, density in an irrigated field reached 80 individuals within a 1 ha trapping grid; the mean (±S.E.) annual density within this habitat was 29.2/ha (±23.5) (Sullivan and Sullivan 2004; Table 1). 

Near Osoyoos and Oliver, up to 22 individuals were found within a 1 ha grid during a single trap session (2 consecutive nights of trapping). Although a formal analysis of these data is not complete, population density estimates likely peaked at ca. 40/ha (W. Klenner, unpubl. data; Table 1).

Alberta

Western harvest mice were the second and third most frequently captured small mammal species during surveys carried out within the Suffield National Wildlife Area during 1994 and 1995, respectively (Reynolds et al. 1999). No mice were captured in 1996. Given that it required three years and 48,578 trap-nights of sampling effort to capture 95 western harvest mice, which represented approximately 5% of the total small mammals captured (95 out of 2,050), it is clear that this mouse occurs at low density in the study area (Reynolds et al. 1999). Between 1982 and 2005, over 6,000 trap-nights along 69 survey transects (14 localities) throughout southeastern Alberta did not yield a single western harvest mouse (D. Gummer, pers. comm.). In addition, several projects involving an analysis of owl pellets collected throughout southern Alberta have identified the remains of several thousand small mammals, none of which were western harvest mice (Schowalter 2004, R. Poulin and R. Schmelzeisen, pers. comm.). These data provide strong evidence for the rarity of this subspecies in Alberta.

Fluctuations and trends

Dramatic seasonal fluctuations have been reported for western harvest mice in BC. During several multi-year studies, population densities have been consistently reported to peak during the fall and winter months and drop off sharply during midsummer (Sullivan and Sullivan 2004, 2005, 2006a, W. Klenner, unpubl. data). A 4-year study did not detect any multi-annual cycle (Sullivan and Sullivan 2005); however, density fluctuated between years (10 to 40/ha; W. Klenner, pers. comm.). Within Alberta, the number of harvest mice captured within the Suffield National Wildlife Area also varied, ranging from 80 in 1994, to 15 in 1995, and zero in 1996. While sampling effort was not equal among years it was of a similar magnitude (Reynolds et al. 1999). 

Rescue effect

Because the western harvest mouse is considered common within the two states that border the BC and Alberta populations (Washington and Montana, respectively; NatureServe 2005, see Existing Protection or Other Status Designations section), the potential for a rescue effect does exist. This potential is enhanced by the dispersal ability (Whitaker and Mumford 1972, Ford 1977) and high rates of recruitment (Sullivan and Sullivan 2006b) reported for this mouse. However, in the event of a significant population decline or local extirpation within Canada, a successful rescue effect may be impeded by the lack of connectivity among suitable habitat fragments located between the secure populations to the south and the potential range in Canada. A better understanding of the habitat requirements of this mouse, as well as its ability to disperse through the existing fragmented landscape of southern BC and Alberta, is required before the potential for a rescue effect should be interpreted as reducing the risk of extirpation in Canada. If the Suffield National Wildlife Area provides the only remaining habitat for harvest mice in Alberta, then a rescue effect is unlikely due to the hundreds of kilometres that separate Montana’s harvest mouse populations from this protected area.