Refining rare weed trait syndromes along arable intensification gradients




How does the conservation (rarity) value of arable weed communities differ along intensification gradients? Which functional traits best distinguish the weed communities of more and less extensively managed fields? Can the same traits predict the rarity of individual weed species?




Using relevé data from 60 cereal and 70 stubble fields, together with weed trait data, we characterized community responses to arable intensification using functional trait analyses based on trait-convergence and trait-divergence assembly patterns. We also examined how well the broad-scale rarity status of species predicts their occurrence along intensification gradients, and how it maps onto our functional classifications describing intensification responses.


The response of weeds to intensification in cereal fields was best described by a functional classification based on species' flowering duration, maximum height and seed weight: weeds of extensively managed fields have short flowering seasons (2–5 mo) and particularly large or small seeds. The highest proportions of rare species also happen to be found in these groups. The rarest weeds among these species tend to be late-winter and early-summer annuals, while the rare species of stubble fields tend to be broad-leaved with low nitrogen requirements, small seeds and short height. Stubble fields showed a decline in weed cover with increasing application of fertilizer and distance from the field edge, but we could detect no strong associations of management factors with trait composition, perhaps because the intensification gradient across these fields was shorter.


Many rare Hungarian weeds are associated with traditional extensive farming practices. They are particularly characterized by short, midsummer flowering periods and by preference for low nitrogen levels, but a range of trait syndromes must be considered to understand their ecology and conservation. Analyses based on trait-divergence patterns, rather than trait-convergence patterns, provide better insights into the functional composition of weed communities, emphasizing the importance of disruptive filters in weed community assembly and the need for improved methods to detect such effects.