Populations of Microtus ochrogaster and Microtus pennsylvanicus were monitored in three habitat types: restored tallgrass prairie (March 1972–May 1997), bluegrass (Poa pratensis; January 1972–May 1997), and alfalfa (Medicago sativa; May 1972–May 1997). M. pennsylvanicus was first observed in the study locations in May 1973.
Study sites
Dry weight of vegetation in the study sites for trapping (clipped at the surface of 26 randomly located 0.25-m2 plots) was used to determine relative abundances of plants.
1. Two restored tallgrass prairies, located in Trelease Prairie (established in 1944) and Phillips Tract respectively. Trelease Prairie was bordered by a mowed lawn, cultivated fields, forbs and shrubs, and a macadam county road, until 1980, when the cultivated field to the south was converted to grazed pasture. The prairies were burned during the spring at 3- to 4-year intervals to set back invading shrubs and trees. Relative abundances of plants:
Trelease Prairie: 17% - big bluestem (Andropogon gerardii), 16% - bush clover (Lespedeza cuneata), 12% - ironweed (Veronia),10% - Indian grass (Sorghastrum nutans), 9% - milkweed (Asclepias), 9% - goldenrod (Solidago), 5% - bluegrass, 5% - switch grass (Panicum), 2% - blackberry (Rubus), 2% - little bluestem (Andropogon scoparius), 1% - about 10 other species.
Phillips Tract site (Sep. 1977): 38% - A. gerardii, 25% - L. cuneata, 16% - Beard tongue foxglove (Penstemon digitalis), 19% - S. nutans (19%), 1% - all other species.
2. The study sites for bluegrass were set up within a former bluegrass pasture in Phillips Tract. The pasture ceased to be used for grazing in June 1971, and by the fall of the same year, it had developed a dense vegetation cover. Bluegrass sites, along with the entire area, were mowed during late summer every 2–3 years, cutting the vegetation at about 25 cm above the surface to suppress growth of invading forbs and woody vegetation. Bluegrass was uncut.
Relative abundances of plants was relatively unchanged throughout 1972-1994: 70% - Bluegrass, 14% - dandelion (Taraxacum officinale), 4% - wild parsnip (Pastinaca sativa), 3% - goatsbeard (Tragopogon), 1% - about 20 other species.
Relative abundance of plants changed between 1995-1997: 75% - brome grass (Bromus inermis), 20% - Timothy (Phleum pratense).
3. Two study sites with alfalfa, separated by a 10-m closely mown strip to limit home ranges of animals whose nests were in one field to extend into the other field. However, animals moved between the two sites and a single population was monitored. Initially, alfalfa comprised 75% of the vegetation in each site. 3-m-wide strips were mowed 25 cm above the surface each summer (early June through mid-September) to control invading weedy forbs and promote new growth of alfalfa plants. Times of mowing were determined so that at least two-thirds of the field was always densely covered by live vegetation and recently mown litter. When alfalfa began to be dominated by invading forbs and grasses on one site, the other side was planted to alfalfa so that the alfalfa would be fully developed when trapping commenced in that site.
Procedures
All study sites were organized on a grid system with 10-m intervals. One wooden multiple-capture live-trap (Burt W. H. 1940) was used at each station. Every month, a 2-day prebaiting period was followed by a 3-day trapping session, except for August 1975 to December 1976 when trapping occurred every 3 weeks at each site. Cracked corn was used both for prebaiting and as bait in the traps. Vegetation or aluminum shields were used to protect traps from the sun during summer, while the wooden traps provided insulation in winter so nesting material was not provided in the traps at any time. Trap mortality was estimated to be <0.5%.
Traps were set in the afternoon. Over the following three days, recordings were taken at approximately 0800 h and 1500 h of: Species, animal, grid station, sex, reproductive condition (males, testes abdominal or descended; females, vulva open or closed, pregnant as determined by palpation, or lactating), and weight to the nearest 1 g. All animals were toe-clipped (<2 toes on each foot) at first capture. The field protocol, which included toe clipping, was in accordance with the guidelines set by the University and Federal authorities, as well as recommendations from the American Society of Mammalogists in effect at that time. The field protocol was reviewed and approved periodically by the Laboratory Animal Resource Committee at the University of Illinois throughout the duration of the study.
The density of voles for each trapping session was estimated using the minimum number known to be alive. Previously marked individuals that were not captured in a given session but were trapped in a later session were considered to have been present during the sessions in which they were not captured. ≤10 individuals must be trapped each session to obtain reasonable estimates. During months that voles were present in the study sites, ≤10 M. ochrogaster were trapped 26%, 52%, and 62% of trapping sessions in alfalfa, bluegrass, and tallgrass, respectively. Ten or fewer M. pennsylvanicus were trapped during 55% of sessions in alfalfa, 46% in bluegrass, and 24% in tallgrass. Because the same index should be used throughout, we felt justified in using MNA. Further, because we used prebaited multiple-capture live traps checked twice daily for 3 days each session, our capture efficiency was very high. Of animals estimated to be present, 92% of the M. ochrogaster and 91% of the M. pennsylvanicus actually were captured each session.
After the end of the 25-year study, trapping according to the protocol above was continued in the alfalfa and bluegrass study sites in October and late November/early December 1997 and 1998 to detect possible population cycles. Trapping was not continued at the Trelease Prairie, because an ecology class that conducted trapping in October of both years yielded very few M. ochrogaster.
