Abstract
The accurate estimate of N2 fixation by legume trees using the 15N natural abundant method (15NNAM) requires the presence of a reference plant, whose δ15N reflects that of the soil derived-N. Our study was conducted in four cocoa (Theobroma cacao L.) plantations located in two cocoa growing regions of Ghana. Gliricidia trees (Gliricidia sepium Jacq. Kunth ex Walp.), a legume which fixes N2 were intercropped in the investigated plantations. The cocoa trees were planted in rows with a density of 3×3 m, while Gliricidia trees were dispersed at variable distances from cocoa (<5 m to >10 m). Our objective was to find out to which extent the presence of Gliricidia in an intercropping system influences δ15N of cocoa (no ability to fix N2) growing at different distances from Gliricidia. We used the 15NNAM which is based on the difference in δ15N values between the soil-mineral N (assessed by the reference plant cocoa) and atmospheric N2, to assess N2-fixation. At the end of both the rainy and dry season, we sampled young, but fully expanded leaves of cocoa trees which were at least 12 m away (referred as ‘cocoa far’) or at 5 m or closer (referred as ‘cocoa close’) to the nearest Gliricidia tree. Gliricidia leaves were sampled as well. Our data show that, except in one site, the δ15N of ‘cocoa far’ was always higher (p<0.05) than ‘cocoa close’. The δ15N of ‘cocoa close’ and Gliricidia were similar in both seasons. Data suggest that ‘cocoa close’ benefited from the N deriving from residues of the Gliricidia tree due to their proximity to this legume tree, and therefore would not serve as a good reference plant for the 15NNAM.