Influence of trees on soil nutrient pools in a silvopastoral system in the Brazilian Savannah

Autor: Rogério Martins Maurício, Iran Borges, Guilherme Lanna Reis, Rodrigo Matta Machado, Talmir Quinzeiro Neto, Ângela Maria Quintão Lana, Regina Maria Quintão Lana
Rok vydání: 2009
Předmět:
Zdroj: Plant and Soil. 329:185-193
ISSN: 1573-5036
0032-079X
Popis: Silvopastoral systems (SPS) have been suggested as an approach to reduce pasture degradation and, consequently, reduce the deforestation of new areas in the Brazilian cerrado (savanna). This study assessed the contribution of trees to soil fertility in a silvopastoral system in Lagoa Santa, Minas Gerais State, Brazil, 19° 35′ 36″ S, 43° 51′ 56″ W; altitude 747 m. The SPS has developed since 1984, through the use of natural regeneration of the native species Zeyheria tuberculosa Vell. Bur., with a density of 160 trees ha−1. The forage component of the system was Brachiaria brizantha cv. Marandu. The control treatment was a pasture adjacent to SPS with the same forage but without the influence of trees. In 2005, the litterfall of trees (leaves, fruits, and stems) was collected monthly, and the dry matter yield and nutrient content were measured. Soil profile samples were collected in February 2006. The litterfall of trees represented considerable inputs of nitrogen (N), potassium (K), and calcium (Ca2+). In the soil, organic matter (OM) and K reduced with depth. At this planting density, this tree species contributed mainly to amelioration of soil acidity, in the 0–2 cm layer and/or in the profile as a whole. In contrast, there were no significant differences in N, P, and K content of the control and SPS soil. This can be related to the high C/N and lignin/N ratios of the litter inputs. Agroforestry systems are complex and the site specific interactions among the components (tree, forage, and animal) must be understood to develop locally adapted systems and optimize productive efficiency. Consequently simple prescriptions for the implementation of SPS by land managers are unlikely.
Databáze: OpenAIRE
Externí odkaz: