Detecting fragmentation of cover in desert grasslands using line intercept

Abstract

Changes in the amount or spatial distribution of grass plants are thought to be indicative of the stability of desert grasslands. This study assessed, through simulation, the sensitivity of statistical properties for distance between plants (fetch length), measured with a line intercept transect, to changes in the spatial distribution and amount of plant cover. Monitoring plots, 30 X 30 m, were simulated for 1, 2.5, 5, 10 and 15% grass cover with random and fragmented spatial distribution. Fetch lengths were measured on 2 randomly placed 30 m transects. In addition to the median and interquartile range, the asymmetry of the sampling distributions was measured with a ratio [(maximum-median)/(median-minimum)] that would identify the presence of at least 1 large open space. The accuracy of the fetch length method was confirmed by the similarity of its sampling distribution to that for the well known random point-to-plant sampling procedure. In both the fetch length and the point-to-plant measures, the median and interquartile range increased with decreasing cover for random and fragmented distribution. The asymmetry estimate increased sharply with increasing cover for the fragmented distribution but asymmetry was nearly constant with increasing cover for the random distribution. The results suggest that the evaluation of changes over time at a monitoring site could use fetch lengths measured along a line intercept transect to detect changes in both absolute and spatial arrangement of cover.