Determine temporal patterns of abundance of rodent and shrew populations and composition of assemblages of small mammals in tallgrass prairie as well as the effects of fire and frequency of fire on these characteristics.
LOCATION OF SAMPLING STATIONS
Ungrazed, unburned - 020B
Grazed, unburned - N20B
Ungrazed, annual burn - 001D
Grazed, annual burn - N01B
Ungrazed, 4 yr. burn - 004B, 004F
Grazed, 4 yr. burn - N04D
FREQUENCY OF SAMPLING
All sites are sampled in late October - early November (after most autumn reproduction has occurred and before winter stress is prominent) and March (before fire on sites to be burned that year) to estimate both early spring density and winter survival (difference in density between autumn and spring).
VARIABLES MEASURED
Numbers of individuals are recorded for each species of small mammal caught on each trap line. Age, sex, reproductive condition, and capture locations of each individual are recorded at each capture. Live mass is recorded at the first capture in each trapping period.
METHODS
Traplines:
Small mammals are trapped on two permanent traplines in each of seven watersheds. Each trapline consists of 20 stations with an inter-station distance of 15 m and terminal stations at least 50 m from the boundary of the watershed. When possible, each trapline was placed so that station 1 was in upland (shallow soil) and station 20 in lowland prairie (deeper soil) and so that the two traplines within a watershed would include 16 stations in upland, 8 stations across limestone outcrops, and 16 stations in lowland. Therefore, the two traplines within a treatment are not truly replicates of each other. Stations 1, 5, 10, 15, and 20 on each trapline are marked with stakes of galvanized steel conduit. All stations are marked with orange plastic surveyor flags before each census.
Trapping Procedures:
Small mammals are trapped for four consecutive days per trapline during each trapping period. Two large Sherman live traps (7.6 by 8.9 by 22.9 cm) are set at each station within 1 m of the station marker. Traps are baited with a mixture of creamy peanut butter and oatmeal (Quaker old-fashioned oatmeal) which is rolled into the shape of a small ball (1.5-2.0 cm in diameter) and wrapped in a 10 cm square of weighing paper. The bait is suspended in the trap by closing the back door of the trap on the twisted end of the weighing paper. Pieces of polyester fiberfill (÷ 5 g), compressed by a #8 rubber band, are used as nesting material in each trap to reduce mortality due to cold weather. With the nest material and a large amount of bait in each trap, mammals are typically in good condition at the time of trap check.
All traps are checked each morning after the end of the nocturnal activity period and before heat or cold stress due to time in trap becomes a problem. Seven traplines are run simultaneously, one in each treatment, followed by the setting of the next series of traplines. Censuses are not conducted during the five day period encompassing full moon conditions.
Animals are marked by clipping a line in the fur with a battery-powered mustache clipper on the right shoulder (first set of traplines in spring census), left shoulder (second set of traplines in spring census), right rump (first set of traplines in autumn census) or left rump (second set of traplines in autumn census) at the time of first capture. This method of marking allows an investigator to count an individual only once within a census line without labor-intensive field and data management of individual identification numbers. Species, sex, reproductive condition, trap station, and any unusual features are recorded at each capture of an individual in each trapping period. Live mass is recorded during the first capture of an individual on a trapline. Reproductive information recorded for males is the presence or absence of scrotal testes. Pregnancy is determined by palpation of the abdomen of females; no effort is made to assess the number of embryos. Presence or absence of conspicuous mammae is also recorded. Individuals are weighed to nearest 0.5 g for those weighing less than 50 g and nearest 1.0 g for those weighing more than 50 g (Pesola balances of appropriate sizes).
FORM OF DATA OUTPUT
The total number of mammals captured by species by line forms the data base CSM04.
SUMMARY OF ALL CHANGES UP TO 1993
In autumn 1981, two traplines were established in each of ten experimental fire treatments (001D, 004B, 004D, 004F, 004G, 010A, 000B, N01D, N04D, and N00D). Live-trap censuses were conducted in spring (March), summer (July), and autumn (late October-November) with animals marked by toe-clipping procedures. Shifts in treatment boundaries were made by the Konza Prairie management during winter 1981-1982, and so two new traplines were established in new treatment N20B in spring 1982. The experimental designation for N01D also was changed to N01B at this time. During spring, summer, and autumn 1982, censuses were conducted using 22 traplines, and then use of the two lines in N00D was discontinued before spring 1983.
From spring 1983 through summer 1984, censuses were conducted using two traplines in each of ten treatments (001D, 004B, 004D, 004F, 004G, 010A, 000B, N01B, N04D, and N00B). In autumn 1984, four new lines were established with two lines in 002C and two in 002D (24 total census lines). Four more trap lines were added in autumn 1985 with two in 010D and two in 001A (28 total census lines). After 1987, summer censuses were discontinued due to the labor required to open and close the traps each day of the census. In 1988, Konza Prairie management changed unburned research treatments to treatments with a 20 year frequency of the occurrence of fire and, therefore, 000B and N00B became 020B and N20B, respectively.
Before spring 1989, the number of trap lines were reduced to 14 with two lines in each of seven experimental treatments (001D, 004B, 004F, 020B, N01B, N04D, and N20B). Selection of the 14 trap- lines was based on the goal of maintaining censuses for small mammals in annual, 4-year, and 20-year burn treatments in both ungrazed prairie (001D, 004B, and 020B) and in the same fire treatments in prairie grazed by bison (N01B, N04D, N20B). In addition, a second ungrazed 4-year fire treatment (004F) was continued to help monitor climatic effects on small mammals in prairie experiencing periodic fires.
In spring 1991, hair-clipping at first capture in each season was exchanged for the more invasive toe-clipping which had been used since 1981. Hair clipping allowed enumeration of the number of different individuals per species caught along each census line which is all that was required for maintenance of the CSM04 data set.