Search for a command to run...
ABSTRACT Background Agri‐environmental measures are used to improve the status of biodiversity and ecosystem services in farming landscapes. Assessing the efficiency of policy implementation relies on species‐ and trait‐based indicators. However, the choice of indicators and analytical approaches to estimate them can affect the interpretation of monitoring results and, consequently, policy design. We assessed long‐term changes in widened grassy field margins and compared the results of three indicator estimation methods. Question Does newly formed vegetation in widened grassy margins advance ecological functionality? What impact does down‐weighting or exclusion of subordinate species have on the interpretation? Data In total, 435 vegetation records from 165 new or widened field margins in Estonia, were observed during 12 years of monitoring. Results Conclusions about vegetation dynamics varied between the indicators and their estimation methods. When an all‐species dataset was used for trait estimation, either as occurrence data or as community‐weighted means, many indicators suggested only minor changes or fluctuating dynamics over time. By addressing abundant species only, evident community changes were detected and many indicators suggested functional improvements. These improvements mostly occurred in the first few years, and the following long‐term changes remained limited. There was general agreement between methods about the decline in potential weediness and the prevalence of taller growth forms with leafy stems. Abundant species indicated improved species richness, support for pollinators and seed foragers. Conclusions Community monitoring should place more attention on abundant species as the main providers of ecosystem services. Information provided by subordinate and rare species is valuable to estimate species richness, but can be noisy to assess ecosystem functional status and mislead to the conclusion of resilience or stability. The slow pace of community change indicates a need for stability in agricultural environmental measures over several policy application cycles.