Comparison of Wheel Point and Point Frame Methods for Plant Cover Measurement of Semiarid and Arid Rangeland Vegetation of New South Wales.

Abstract

Ground cover is frequently estimated in rangeland monitoring and it is an important intermediate measurement between biomass estimation and satellite imagery. As a preliminary phase in a longer term program, wheel point and point frame methods were used to measure vegetation cover on four permanent Soil Conservation Service transects at each of four land systems in western New South Wales, at Nyngan (410 mm average annual rainfall), at Cobar (364 mm average annual rainfall) and two at Fowlers Gap (200 mm average annual rainfall) north of Broken Hill. The majority of this work used 400 wheel point hits per transect and 100 point quadrats sub sampled 9 - 13 times along each of four transects. There was no statistically significant difference between these techniques for total foliage cover over a combined analysis of all sites under pre-drought conditions, and for pre-drought and post-drought at Cobar. However, there was a 10% difference estimated between the techniques for total foliage cover at Nyngan when it was analysed in isolation. There were no consistent differences in technique for cover estimation for more than 40 plant species including annual grasses and herbs, perennial grasses and saltbushes. Significant differences between techniques were found for Medicago sp. and Thyridolepis mitchelliana on one occasion. We believe that these differences were due to the problems of finding small plants in tall grass and identifying heavily grazed grasses during drought conditions at Cobar and, in the latter case, this was also associated with a significantly greater estimate of mean cover for all grasses and thus total foliage cover. Although there was generally no statistical difference between techniques our observations suggest that the point frame tends to give lower estimates of cover than the wheel point in the situations measured. This may be associated with the circumference of the marker pins on the wheel point or perhaps observer error but as this effect appeared to be more noticeable with grasses we suspect that the former is most likely. The wheel point is less time consuming, more convenient and simpler to use than the point frame, and will readily accommodate most temporal and spatial variation in sampling requirements in similar land forms in western New South Wales.