The aims of this dissertation were to quantify and to modelling the nutrients in aboveground
biomass for Mimosa scabrella Bentham under different site, age and
diameter classes. The data for this research came from 25 trees sampled into native
Mimosa scabrella stands of the northern Curitiba metropolitan region, Paraná State,
Brazil. The trees were cut down and their components were separated in: leaves,
twigs, branches, stem wood and bark. Then, a sample of each component was taken
to obtain dry matter. Later on, chemical analysis was proceeded to determine N, P,
K, Ca, Mg, Fe, Mn, Cu, Zn concentrations for each tree component. To evaluate tree
component, site, age and diameter effects over nutrient concentrations and nutrient
contents, a completely random experimental design was used, where the statistical
analysis was constituted by Bartlett’s test, analysis of variance and Tukey’s test.
Several mathematical models were fitted intending to obtain nutrient content
estimates for different tree components using tree dimensions, being them: traditional
volume models found in the literature, mathematical models specifically built for this
work and built by stepwise procedure. The results found in this research indicated
that the major of nutrient concentrations presented a decreasing order relative to
different tree component as follow: leaves > stem bark > twigs > branches > stem
wood. In a general way, the decreasing order of nutrient concentration for the major
of tree components was: N > Ca > K > Mg > P > Mn > Fe > Cu > Zn. Considering
stie, age and diameter effects over nutrient concentrations, only the site classes
showed a significant and coherent influence over some nutrient concentrations. It
was verified that P and K leaf concentrations increased with site quality. The nutrient
content of the stem wood presented the largest quantity of nutrients, but if firewood of
Mimosa scabrella was considered, it owns 70% of total above-ground biomass
nutrient content. It turns clear the great Mimosa scabrella stands potential of nutrient
losses at crop (harvest). For the nutrient content in the tree components, only the
diameter classes presented a significant and coherent influence. The nutrient content
increases with the diameter classes. Perhaps, better results of the the site, age and
diameter effects over nutrient concentration and content could be found if more trees
and stands had been sampled, or with a severe control of the site, age and diameter
interrelationships. In a general way, the macronutrient contents were well correlated
with tree dimensions, and acceptable equations were derived. However, the same
did not occur to the micronutrients, except for Cu and Fe for some tree components.
Generally, nutrient content equations presented better performances to the bole
components and to the added tree components (leaves + twigs, firewood and total
above-ground biomass). In some cases, the insert of bole and crown dimensions,
age and dominant height as independent variables improve the nutrient content
equations.