Miguel F.Acevedo
ABSTRACTS
Growth dynamics of three tropical
savanna grass species: an individual module model
Ecological
Modelling. 154:45-60. 2002.
Miguel
F. Acevedo1 and
Josep Raventós2
1Institute of Applied
Sciences and Department of Geography, University of North Texas,
Denton, Texas
76203.
2Departamento de Ecología,
Universidad de Alicante, Alicante, España.
Abstract
We model the dynamics of grass plant growth as a collection of the
individual dynamic behavior of shoots and leaves. The model is inspired
in data for plants of three species (Elyonurus adustus, Leptocoryphium
lanatum and Andropogon semiberbis) of common grasses in the Venezuelan
savannas that were sampled monthly for one year. These species
represent different architecture and regeneration response to fire.
Modules (shoots and leaves) were counted in each cell of a square grid
in each one of several vertical levels. Module density per cell
provides the horizontal distribution within a level and is aggregated
by level to obtain vertical distribution. Both distributions are
simulated by a dynamical model based on shoot emergence and mortality,
elongation of shoots and leaves given by Richards’ equation, plus a few
simple geometric considerations. For quantitative comparisons of model
results to data, the transient and final values for vertical
distribution plus two metrics of horizontal distribution at each level,
were calculated for the simulation results and the field data.
Proportion of occupied cells and maximum distance to the center of
growth were the two metrics selected to capture the dispersion and
range of the horizontal distribution. The model results indicate
predictable final vertical profiles (of proportional density plus the
two metrics) similar to the profiles of measured distributions for each
species. A reasonable prediction of the transient behavior was also
obtained but with larger deviations as evaluated by the root mean
square error between model and data. Differences in vertical and
horizontal patterns of module density among species are explained by
changing a set of parameter values related to growth form and
phenology. Thus, the model could be applied to generate plant
functional types for analysis of savanna dynamics subject to fire. With
modifications, the model is potentially applicable to other grass
plants and other grassland ecosystems.
Keywords: Growth, phenology, shoots,
Richards’ equation, savanna, grasses, Venezuela, fire.
Back to list