Decomposition trials provide useful insights in certain properties of woody
species. The Figures 1 and 2 effectively demand a clarification of management
priorities: "Do you need tree species efficient in accumulating potassium
since you are searching for alley-cropping species for potassium-poor
environments, or, is lasting soilcover for water conservation the issue because
of unreliable rainfall?" In the first case Leucaena leucocephala and
Gliricidia sepium will be obvious candidates, in the latter situation Flemingia
macrophylla is more likely to succeed.
Whatever is the case, acquiring knowledge on the decomposition characteristics
of the leaf biomass will be part of any systematic approach to an
inventory and subsequent screening of species for use in alley-cropping
systems.
Such approach will gain in efficiency if a standard methodology is adopted.
Considerations in this respect are to include a detailed description of
climatic conditions during trials, particularly information on rainfall and
temperature.
Secondly, the methodology applied in decomposition studies should be
evaluated. Objections can be made to results of decomposition studies in
which prunings are sealed in bags made of nylon mesh, and left to decompose
[17]. This method excludes a whole range of decomposing agents, such
as larger insect species. Moreover, the use of bags shields mulch layers from
the direct, destructive impact of raindrops during heavy showers. As the
data in this study show soil-dwelling fauna potentially is a factor explaining
loss patterns in mulch layers.
A third suggestion is to adopt a certain mulch type, for example that of
Leucaena leucocephala, as a standard to interpret the results of comparative
mulch decomposition studies. Such procedure is for example accepted in
animal feed digestibility studies, and would be of help in comparing results
from different sites and materials.
Eventually, alley-cropping studies will enter the field of modelling. It is
simply inconceivable that the large bodies of data that now start to flow
from inventories can be handled without the help of computer models. One
future objective will be to simulate alley-cropping systems in order to predict
and optimise their relative efficiency. In the complex exercise of grasping the
dynamics of alley-cropping systems, decomposition- and nutrient release
rates of mulches will be found as simple but essential parameters in models.