Monitoring the response of vegetation to use by large herbivores: an assessment of some techniques

The suitability of two vegetation survey techniques (point surveys of canopy spread cover, and counts of individual plants in quadrants) was assessed. The cover estimates were subject to between-observer differences which could not be eliminated despite clearly defined rules governing the scoring of strikes. The magnitude of these differences was slight, but nevertheless operator bias would need to be checked if trends in cover were being monitored. Variation between replicate point surveys conducted by one observer was also assessed. The point surveys contacted no more than 60-70 % of plant species known to be present in the survey plots. This technique is therefore likely to underestimate plant species richness. As a result of improved rainfall grass tuft density showed a five-fold increase from November 1984 to December 1985. Grass canopy spread cover also showed a pronounced increase over this period. The implications of climate induced changes for the design of vegetation-monitoring programmes are discussed.