Modelling changing sensitivity to alternating temperatures during induction of secondary dormancy in buried Polygonum aviculare L. seeds to aid in managing seedbank behaviour

Abstract

Seed dormancy is a common attribute of many weed species in temperate habitats that controls the seasonality of weed emergence from soil seedbanks. For some weed species, it is not enough to reduce seed dormancy, as germination can only proceed after exposure to dormancy termination factors (i.e. light and alternating temperatures). Sensitivity to alternating temperatures in Polygonum aviculare seeds increases during dormancy release. However, it is not known whether this sensitivity is lost during dormancy induction. The aim of this study was to investigate and quantify the changes in dormancy level of P. aviculare seeds during secondary dormancy induction, as measured by changes in sensitivity to alternating temperatures driven by the soil temperature. To achieve these objectives, seeds were stratified at 5°C until obtaining a minimum dormancy level. The seeds were then buried in pots and induced into secondary dormancy by storing them at 10, 15, 20 and 25°C for different time periods. During storage, seeds were exhumed periodically and exposed to different cycle-doses of 10/24°C (12 h/12 h) (0, 1, 3, 5 and 15 cycles) to test germination. Our results showed that the high sensitivity to alternating temperatures achieved during primary dormancy release, decreased during secondary dormancy induction with a rate that is temperature dependent. These changes in sensitivity to alternating temperatures were quantified as a function of the accumulation of thermal time, which was calculated over a base temperature of 7.9°C. Coupling the present model with a previous one to assess changes in sensitivity to alternating temperatures during dormancy release, allowed us to develop a model for the prediction of cyclic changes in sensitivity to alternating temperature in relation to the thermal environment experienced by the seeds during burial. We present this model as a valuable tool for managing the weed seedbank through different agronomic practices.