Carbon and Nitrogen Cycling in a Tree - Grass Inter-Cropping System in the Humid Tropics of Mexico.
This work aimed to contribute to the understanding of tree - grass inter-cropping interactions so that the productivity and sustainability of extensive livestock husbandry can be increased. The work was carried out in the context of a small farm in Oaxaca, Mexico, where increases in productivity are limited by shortage of capital and where the tree component would be used as green manure. It is difficult to investigate the effectiveness of such a system by only using conventional field trials. I constructed a mathematical model to simulate how the main components of the system function under conditions that would not be evaluated in the field. Issues such as how many trees to plant and what tree species combine with grass cattle and environment, can be answered with the model. The particular features of the model are: 1) It describes an agro-ecosystem where trees perform several biological functions like nitrogen capture for use in the silvopastoral system, 2) It links grass and trees with the animal and 3) Nutrient availability depends mainly on soil organic matter decomposition and mineralisation rather than on external inputs. The present research consisted of 1) constructing the model prototype using data from the literature, 2) conducting field experiments to investigate the actual performance of the silvopastoral system, 3) perform laboratory research and greenhouse experiments complementarily to the field experiments and 4) elaborate on the carbon and nitrogen balance of the silvopastoral experiment, by combining research results and the mathematical model. The field experiment consisted of an array of 13 plots with one of the tree species Gliricidia sepium, Leucaena leucocephala, Delonix regia and Lysiloma auritum in a gradient of plant densities within a Brachiaria decumbens paddock. Results showed that the presence of trees in pastures is potentially useful for retaining nitrogen and carbon that would be lost in the grass mono-crop. Trees did not incorporate nitrogen through biological fixation, perhaps because the lack of adequate nodulation and they did not established their rooting systems to a depth beyond the grass roots (> 1.20m) so as to recover leached nutrients. However, trees produced mulch that was rich in nitrogen (3.8%) and whose decomposition rate ensures a slow release to prevent leaching. At the plant density used, the tree population caused no harm to grass as to production and nutritive value. Further increments in tree density in order to improve the potential for nitrogen capture should be evaluated in terms of the reduction of grass production. Several biological attributes of the species were determined, in some cases for the first time: biomass productivity, specific leaf area, nutritive value, phenolic content, root biomass, grass root longevity, root vertical distribution, etc. Such characterisation is useful for the understanding of the system inter-cropping and specially for the parameterisation of the silvopastoral model. Even though the mixtures proved able to survive for the span of the experiment, the sustainability of tree - grass inter-cropping as to the stabilisation of soil fertility requires longer monitoring. Other limiting factors such as phosphorus availability and the management of grazing systems have to be incorporated for an adequate evaluation of the silvopastoral system.