Submitted to: Range Management Meeting Proceedings
Publication Type: Abstract only
Publication Acceptance Date: 10/1/2001
Publication Date: 1/1/2002
Citation: NORTHUP, B.K. SPATIAL VARIABILITY IN THE PRODUCTION OF HERBACEOUS BIOMASS BY A SOUTHERN TALLGRASS PRAIRIE IN OKLAHOMA. RANGE MANAGEMENT MEETING PROCEEDINGS. 2002. Abstract. p. 189. Interpretive Summary: Abstract Only.
Technical Abstract: Herbaceous characteristics of plant communities in semi-arid areas can be spatially variable, which can affect the capacity of sampling techniques to define the productivity of landscapes. This study was undertaken to describe variability in distribution of herbaceous biomass produced by a southern tallgrass prairie site in central Oklahoma, and its potential impacts on sampling forage production. The study was conducted on a 0.45 ha (30 x 150 m) plot dominated by tallgrass species, with a northwesterly exposure and situated between a ridge and stream bank. Random locations (n=30) were identified along seven, 150 m transects (running from ridge to bottom of slope) spaced 5.0 m apart. Herbaceous materials were collected from 0.25 m2 quadrats at locations during mid-June and August 2001, the predominant species (> 70% of production) identified, and elevations measured. Spatial variability in production was described by geostatistical analyses. Neighboring pairs of observations were spatially auto-correlated across distances < 30 m parallel to the slope, and < 18 m perpendicular to the slope. These scales of auto-correlation defined the size, shape and distribution of different patches within the plot. Highly productive zones (top third of production range) within this plot were restricted to small areas; 35% of the plot was comprised of scattered patches containing big bluestem (Andropogon gerardii), indiangrass (Sorghastrum nutans) or switchgrass (Panicum virgatum). Low production (bottom third of range) patches (50% of the plot) were dominated by dropseed (Sporobolus asper) or little bluestem (Schizachyrium scoparium). Samples collected to define mean production, or treatment responses, on this site should be about 30 m apart to be considered independent.