Effects of herbicide and nitrogen fertilizer on non-target plant reproduction and indirect effects on pollination in Tanacetum vulgare (Asteraceae)

Abstract

Herbicides and nitrogen fertilizers are widely used in intensively grown agricultural areas. Non-target plants growing in habitats adjacent to conventional fields may be exposed to herbicides by spray drift and misplacement of nitrogen fertilizers. Whereas sub-lethal effects of herbicides have previously been documented, combined effects of nitrogen and sub-lethal herbicide exposure are less well known. Moreover, indirect effects on pollination and fruit set following effects of agrochemicals on flowering remain largely unexplored. In the current study, we investigate combined effects of herbicide and nitrogen fertilizer on reproductive features of Tanacetum vulgare (Asteraceae). The study was carried out in an experimental set-up, in which plots of 7 × 7 m were treated with one of six treatments: four levels of the herbicide glyphosate (0%, 1%, 5%, and 25% of label rate of 1440 g a.i. ha−1) without nitrogen in addition to two levels of herbicide (0% and 25%) with added nitrogen (100 kg N ha−1 year−1). The set-up had a randomized full-factorial design with 10 replicates of each treatment. We monitored floral density per plot every fortnight during flowering, stem height at flowering, diameter of capitula, visitation rate and diversity of flower-visitors, in addition to weight per seed and total seed weight per seed head. Glyphosate, nitrogen and their interaction affected plant reproductive features, in particular, floral density and flowering phenology. Glyphosate negatively affected floral density, and flowering was severely delayed by glyphosate application (10.5 days delay per 100 g a.i. ha−1 year−1). Although nitrogen partly mitigated reduction in floral abundance, the delay in flowering was amplified when nitrogen was added. Effects of glyphosate on flowering were mirrored in reduced flower visitation by insects and reduced seed set. These effects occurred even when plants were exposed to glyphosate two months before onset of flowering. Our results show that misplacement of agro-chemicals markedly changes floral abundance and flowering phenology of non-target plants, which may secondarily impact pollination interactions and reproduction of wild plants in agricultural landscapes.