The study was conducted on the Konza Prairie Research Natural Area located about 15 km south of Manhattan, Kansas. Vegetation of this area is typical for tallgrass prairie and was dominated by big bluestem (Andropogon gerardii Vitman), Indiangrass (Sorghastrum nutans (L) Nash) and switchgrass (Panicum virgatum L). Further details of vegetation composition are reported in Bragg and Hulbert (1976) and Hulbert (1969). The site used in this study was a bottomland area on a watershed that had not been burned for three years. No grazing by cattle had occurred on this site for 10 years. Soils were fine mixed mesic pachic argiustolls formed in colluvial and alluvial sediments and have been described in Jantz et al. (1975).

A series of 18, 10 m X 10 m plots was established parallel to stream-bank. Treatments (unburned, burned, mowed, and raked three times during growing season) were blocked to remove the variants that might be due to slope effects. The study was initiated in the Spring of 1981, and treatments were continued for two years. Soil arthropods were censused by excavating and hand-sorting two or three 0.1 m² by 20 cm deep soil monoliths per plot (Seastedt 1984a). Rhizomes of grasses were concurrently harvested from these samples and sorted into living and dead components. Arthropods were returned to the laboratory, identified, dried at 70°C and weighed. Samples were obtained when the arthropods were inactive during November 1981 and again in late March and early April of 1983.

Arthropods were identified to family and grouped according to trophic status. Only larger arthropods (macroarthropods) are reported here. Ants and termites were not included in the counts, nor were densities of the very small arthropods (microarthropods). All white grubs (Scarabaeidae) were classified as herbivores; few adult detritivores of this family have been found of ungrazed prairie (Seastedt 1984a).

Most arthropod groups censused in this study exhibited clumped distributions. Therefore, a logarithmic transformation (y=log(x + 1)) was performed prior to statistical analysis to normalize distributions and homogenize variances. Differences in densities and biomass of arthropods attributed to sampling date were assessed by using a two-way ANOVA (date, treatment and interaction) using type-IV sums of squares to establish the potential significance of sampling date. Treatment effects on arthropod densities and biomass were evaluated for each collection date using one-way ANOVA.

Kucera and Dahlman (1968) established a positive correlation between rhizome and root mass and productivity. Living and dead rhizome mass were therefore used as indices of belowground plant biomass and plant productivity on the treatments. Big bluestem and Indiangrass dominated total rhizome biomass, and only these two species were used in our index of belowground plant productivity. Other indices of plant productivity and nitrogen content of belowground plant materials were obtained from other studies being conducted on Konza Prairie or nearby areas by other investigators.

For additional metadata information see: http://lter.konza.ksu.edu/sites/default/files/DC.pdf

For additional methods information see: http://lter.konza.ksu.edu/sites/default/files/MM.pdf