Recovery from autumn defoliation of spring sown tagasaste and non destructive techniques to estimate yield : A dissertation submitted in partial fulfilment of the requirements for the degree of Agricultural Science in the University of Canterbury

Abstract

From March to July of 1984 three experiments were conducted to investigate, (1) the regrowth of autumn grazed and cut tagasaste (Chamaecytisus palmensis), (2) the influence of autumn climate on the growth of tagasaste, and (3) non destructive techniques for estimating dry matter yield of tagasaste. In the first experiment plants were subjected to the following treatments in March, uncut, lightly cut, severely cut, sheep grazed, and simulated grazing. The defoliation treatment that yielded the most regrowth up till July was light cutting. It produced 34.4 g DM per plant more than the simulated grazed plants. The simulated grazed plants behaved similarly to the sheep grazed plants, which were severely defoliated, and showed poor recov.eries of 15.5 and 19.5 g DM respectively. A light autumn grazing of spring sown tagasaste would not be detrimental to subsequent regrowth. In the second experiment, growth rate and accumulative growth decreased as sunshine hours and temperatures declined. The best accumulative growth and growth rate from the defoliation treatment was obtained from quaternary laterals attached to cut tertiary branches. The final accumulative growth achieved was 710 mm in length for laterals attached to cut branches compared to 347 mm for laterals attached to uncut branches. The maximum growth rate achieved from quaternary laterals attached to cut branches was 71 mm/week and 30 mm/week for those attached to uncut branches during a similar time period. In the third experiment, the best non destructive technique used for estimating browse yield was with tagasaste photographed against a ·'graduated background (r² = 96.3%). Visual scoring was the next best technique that gave a good correlation with total and edible browse yield (r² = 90.2%, r² = 83.3% respectively). It would provide a good backup estimate to using a graduated backdrop and also for rapid assessment in a farming system. Stem diameter, as a single plant factor, was the least successful (total yield r² = 65.8%, edible yield r² = 64.2%), since it was not a sensitive parameter in estimating yield of various defoliation severities.