A influência dos métodos de controle e dos níveis de adubação na composição de plantas infestantes na restauração florestal

Abstract

Weed species, especially those from the Poaceae and Cyperaceae families, cause several problems in the first stages of tree seedling plantings aiming for tropical forest restoration. The general objective of this work is to evaluate the diversity and biomass of weeds in forest restoration plantations submitted to different weed control techniques and fertilization intensities. Do species composition and richness vary depending on weed control techniques and fertilization levels? Does the biomass of unwanted herbaceous plants vary as a function of weed control techniques and fertilization levels? When we intensify the control of weeds with the use of herbicide, does the proportion of biomass and the richness of non-monocots weeds increase? The hypothesis was that in the treatment carried out with intensive management of weeds (glyphosate spraying and mechanical weed control around seedlings every three months) we would have a lower biomass of Poaceae and Cyperaceae members, and a greater overall richness of weed species. On the other hand, the treatment with conventional management (mowing and weed control around the seedling every six months) and with the use of fertilization, would have a higher biomass of Poaceae and Cyperaceae species, and a lower weed species richness overall. To answer these questions, two management techniques were used to control weeds: intensive management with seedling crowning; and conventional management using traditional mechanized mowing and seedling crowning. Fertilization was carried out two times, first in holes and then over the soil in a circle, both place around the seedlings. In the first one, right after planting, we applied 85 g per seedling of the NPK 14-04-08 fertilizer. In the second moment, 45 g per seedling of the NPK 10-0-20 was applied. A full randomized design in the “Split-plot” scheme was used. For the weed inventory, we used 0.25 m2 squares, placed both in the row (around the seedlings) and in the interow. We used ANOVA followed by the Tukey´s test for the statistical analyses. We found 38 weed species, but found neither effect of fertilization and control methods nor interaction among them on the measured variables. At the same time, there was a higher biomass for monocots and in the interow of the plantings. Placing some samples in the row, around the seedlings, may probably favored the unexpected results. The study shows us that knowledge on the diversity and biomass of weeds in area to be restored can help deciding on the best control methods to be used in forest restoration projects.