SOME CONSIDERATIONS FOR THE USE OF VISUAL ESTIMATES OF PLANT COVER IN BIOMONITORING

SUMMARY (1) The magnitude of changes required to detect air pollution effects was determined by an error analysis of visual estimates of plant cover in a Pinus contorta var. latifolia stand in western Canada. (2) Species with the largest measurement errors had the smallest covers. Factors that appeared to contribute to the errors included plant morphology, distribution and the incorrect identification of species. (3) The observer's familiarity with the vegetation enhanced measurement precision but only for the short term. (4) Removal of species with measurement errors of 20% had little effect on the similarity of repetitive measurements. (5) The time of sampling influenced the results. Sampling between mid-June and mid-August, when there was minimal change in the vegetation, virtually eliminated this source of error. (6) Changes in the vegetation need to be >20% before they can be attributed to factors other than annual fluctuation and measurement error. This uncertainty may be a limiting factor in the utilization of plant cover estimates for bio-monitoring purposes.