Optimized Frequency Measures for Monitoring Trends in Tallgrass Prairie

Abstract

Prairies exhibit a bimodal frequency distribution of patch occupancy with a few dominant species accounting for most of the primary production and plant biomass and numerous other species occurring infrequently over time and space. Consequently, management interest is often focused on the abundance of common species, and on the persistence of less common species. We used spatially nested arrays of various sized plots to simultaneously measure species’ persistence and abundance at Tallgrass Prairie National Preserve, Kansas. Larger plots captured more species, with 127 species identified at the 10-m scale. Smaller plots, however, provided useful frequency data for measuring the abundance of common species. We defined a plot frame size that delivers a frequency in a 20–50% target range as the optimal plot size, and the frequency of a species sampled at its optimal plot size is termed the optimized frequency. Optimized frequency was estimated for 27 common species for the years 2002– 2006. Of the nine perennial grasses in this group, five exhibited significant interannual variation within the 5-yr period. Bouteloua curtipendula exhibited a significant linear change over the 5-yr period, decreasing in frequency over time. From year to year, changes in optimized frequency among the 27 common species varied in both their trajectory (i.e., the average change), and amplitude (i.e., the average magnitude of change). Between 2005 and 2006, the average change in optimized frequency was 29.1%, the only year with an overall declining trajectory. The declining trajectory between 2005 and 2006 was significantly different from the increasing trajectories observed between 2002 and 2003 and between 2004 and 2005. No significant differences were observed in amplitude. We offer these techniques for aggregating information from multiple spatial scales to characterize the amplitude and trajectory of community-wide changes as tools to help scientists convey the results of multiscale research projects to land managers.