Stochastic population model of Zea mays L.

Abstract

We propose a minimalist stochastic population model of maize, focused on the description of the maize vegetative stages (seedlings with different number of leaves) involved in the propagation of vector-borne diseases. This model was parameterized from laboratory and field experiments and from observational field studies for multiple hybrids and different weather and soil conditions, taking into account only temperature as input variable. We propose three different submodels to estimate the distribution of the Final Leaf Number N FLN in the plants and to estimate the tassel initiation probability. The first submodel (submodel A), with a fixed N FLN , is adaptable to any particular hybrid, the second and third submodels allow to simulate plants with an empirical N FLN distribution according to bibliographic averages (submodel B) or according to a Poisson Process (submodel C). The three submodels are able to describe the temporal development of populations and events. A good agreement is observed between the development times predicted by the model and the values obtained from laboratory experiments at constant temperature, field experiments carried out in Brazil and Australia and observational studies performed in Argentina. This model may be improved and coupled to leaf growth models and leaf area estimation models to be able to estimate not only the temporal development of populations and events but also the temporal development of the leaf area by plant, which is believed to be related to the carrying capacity of maize specialists insects, vectors of maize diseases.