Simulation of Pollen Transport by Wind

The authors present a Cellular Automaton (CA) model designed to simulate how wind transports and deposits pollen. The model uses a 2D lattice to track pollen through three states: releasable, air-borne, and deposited. It incorporates stochastic processes for pollen release, transport (including turbulence and gusts), and 'sticking' probabilities for deposition. The authors demonstrate that this CA approach provides a better fit for experimental pollen deposition data than traditional exponential decay models and can reproduce complex patterns like humps or irregular declines. The paper also validates the model by showing it can replicate the Gaussian diffusion patterns predicted by partial differential equations for a thin line of pollen moving in a uniform wind.

• The authors argue that a simple exponential law is a poor approximation for pollen deposition over distance because it fails to account for the non-stationary nature of wind and weather conditions. Comparison with data from Jones and Brooks (1950) shows that observed curves have a steeper slope at small distances and a slower decay in the tail than the best-fit exponential decline.
• The developed Cellular Automaton (CA) model can provide a more accurate fit to field experiment observations than exponential decline models, provided the parameters are suitably chosen. The model is capable of producing decay curves with 'humps' or irregular declines, which are sometimes observed in real-world data.
• The CA model represents space as a regular lattice of square cells (typically 10 to 50 metres per side) and tracks pollen through three states: 'created and releasable' (Ac), 'free and air-borne' (Af), and 'sticking and deposited' (As).
• Pollen transport is simulated stochastically using four wind components: a time- and space-independent component, a wavelike propagating component, a localised component (gusts), and a randomly fluctuating component (turbulence). An 'entrainment factor' (fv) is used to account for the fact that pollen typically travels slower than the actual local wind speed.
• The model can be extended to three dimensions to incorporate high-altitude winds (above approximately 500 m). This is achieved by adding a fourth pollen state ('high-altitude pollen') and using a layered system where vertical motion is managed by exchange probabilities between the ground, low-altitude atmosphere, and high-altitude layers.
• Numerical experiments involving a single column of plants under uniform wind demonstrate that the CA model reproduces a Gaussian diffusion profile. This result matches the analytical solution of a partial differential equation, validating the CA model's ability to simulate the random-walk diffusion caused by small-scale velocity fluctuations.

APA

Novotny, E., & Perdang, J. (n.d.). Simulation of Pollen Transport by Wind. Scientists for Global Responsibility. https://www.sgr.org.uk/sites/default/files/PollenTransport-Paper I-final-SGR.pdf

Chicago

Novotny, E, and J. Perdang. Simulation of Pollen Transport by Wind. Scientists for Global Responsibility, n.d. https://www.sgr.org.uk/sites/default/files/PollenTransport-Paper I-final-SGR.pdf.

Wikipedia

{{cite report |last1=Novotny |first1=E |last2=Perdang |first2=J. |title=Simulation of Pollen Transport by Wind |publisher=Scientists for Global Responsibility |url=https://www.sgr.org.uk/sites/default/files/PollenTransport-Paper I-final-SGR.pdf |access-date=23 October 2026 |via=Climate Insights Directory}}

BibTeX

@techreport{novotnyndsimulation, author = {Novotny, E and Perdang, J.}, title = {{Simulation of Pollen Transport by Wind}}, institution = {Scientists for Global Responsibility}, url = {https://www.sgr.org.uk/sites/default/files/PollenTransport-Paper I-final-SGR.pdf}, urldate = {2026-10-23}, note = {Indexed by Climate Insights Directory} }