Biophysical characteristics of maize pollen: Variability during emission and consequences on cross-pollination risks

Modelling of cross-pollination risks using physical-based model of pollen dispersal requires estimations of pollen viability rate as well as settling velocity distribution of viable pollen grains at emission. These two biophysical characteristics are mainly determined by pollen water content. Three years of experiments were carried out to measure biophysical characteristics of pollen at different times of emission within the day. In a first part, we proposed a simple model to predict pollen water content at emission as a function of air vapour pressure deficit. We found that by setting a constant duration of pollen dehydration in the anthers the model predicts accurately pollen water content variations at emission. In a second part, we related pollen water content and settling velocity distribution of viable and non-viable pollen grains. We used with success settling velocity distribution as a proxy to estimate fraction of spheroid pollen, water content distribution of spheroid pollen and pollen viability rate. In addition, we validated with success the estimations of viability rate. In a third part, based on above models, we estimated the effects of air humidity on cross-pollination risks.

► Pollen water content at emission decreases over a day due to decreasing air humidity. ► Pollen dehydration induces a decrease of viability rate and settling velocity. ► Bimodal distribution of settling velocity results from spherical or prismatic shapes. ► Settling velocity threshold of 23.54 cm s−1 segregates viable and non-viable pollen. ► Settling velocity could be used as a proxy to estimate viability rate.