Mechanistic analytical models for long-distance seed dispersal by wind

Long-distance dispersal (LDD) of wind-borne organisms is central to quantifying risk for transgenic escape and gene flow, control of pests and invasions, persistence in fragmented landscapes and species co-existence; yet LDD remains notoriously difficult to define, measure and model. This difficulty has shaped the current paradigm that the frequency and spatial extent of LDD events are almost impossible to predict.

In the September issue of The American Naturalist, G. G. Katul (Duke University) and colleagues introduce a mechanistic analytical model for estimating dispersal kernels of seeds and their escape probability from the canopy, using simplifications to well-established turbulent transport theories. The model parameters–wind statistics, seed release height, and seed terminal velocity–are clearly interpretable and can easily be measured independently of dispersal data, as compared to the synthetic parameters of equivalent phenomenological analytical models that necessitate dispersal data for calibration.

A necessary condition for LDD, seed uplifting and escape from the canopy, along with other key attributes of the dispersal kernel, were reproduced well by the model. To meet the increasing demand for proper evaluation of ecological risk reduction by employing less subjective and more transparent methods, mathematical models should make their assumptions explicit and should realistically incorporate the key biological and physical processes underlying environmental changes.