E. Salas et al., Estimating biological N-2 fixation by a tropical legume tree using the non-nodulating phenophase as the reference in the N-15 natural abundance method, SOIL BIOL B, 33(12-13), 2001, pp. 1859-1868
The non-nodulating phenophase of a legume tree was tested as a non-N-2-fixi
ng reference in application of the N-15 natural abundance method for estima
ting the N-2 fixation. We applied this method to study the effects of three
pruning intensities, complete pruning every 6 months (T-6), ca. 50% prunin
g every 3 months (P-3) and intact control (C), on N-2 fixation in Erythrina
lanceolata (Papilionaceae) planted as shade and support trees for vanilla
(Vanilla planifolia) in a subhumid tropical site in Quepos, Costa Rica. We
measured nodulation and N2 fixation for 12 months. The trees under the C re
gime nodulated abundantly during the rainy season vegetative growth but did
not nodulate during the dry season and flowering. A linear regression (r(2
) = 0.76) was observed between the nodule biomass and ON values under the C
regime, suggesting quite a stable specific N-2 fixation rate. The stable d
elta N-15 values throughout the year in the non-N-2-fixing Morus nigra (Mor
aceae) growing on the same soil indicated that the temporal variation in th
e plant available soil N-15 was low. We used the intercept term of the regr
ession (delta N-15 = 3.5 when nodule biomass was 0) as the non-N-2-fixing d
elta N-15 value when estimating the percentage of N fixed from atmosphere o
ut of tree total N. The percentage varied from 0 during the driest period t
o 53% during rainy season. Pruning reduced rainy season nodulation under th
e T-6 and P-3 regimes almost to nil, and the delta N-15 values were high. O
ur results suggest that the conditions for using the non-nodulating phenoph
ase as the non-N-2-fixing reference required in the N-15 natural abundance
method were fulfilled. The C trees showed a clear phenological cycle in N-2
fixation, while pruning severely disturbed the N-2 fixation under the T-6
and P-3 regimes, indicating that E. lanceolata is better suited for agrofor
estry systems in which the trees are not managed by periodic prunings. (C)
2001 Elsevier Science Ltd. All rights reserved.