Over the last 30 years, many studies have surveyed weed vegetation on arable land. The ‘Arable Weeds and Management in Europe’ (AWME) database is a collection of 36 of these surveys and the associated management data. Here, we review the challenges associated with combining disparate datasets and explore some of the opportunities for future research that present themselves thanks to the AWME database. We present three case studies repeating previously published national scale analyses with data from a larger spatial extent. The case studies, originally done in France, Germany and the UK, explore various aspects of weed ecology (community composition, management and environmental effects and within-field distributions) and use a range of statistical techniques (canonical correspondence analysis, redundancy analysis and generalised linear mixed models) to demonstrate the utility and versatility of the AWME database. We demonstrate that (i) the standardisation of abundance data to a common measure, before the analysis of the combined dataset, has little impact on the outcome of the analyses, (ii) the increased extent of environmental or management gradients allows for greater confidence in conclusions and (iii) the main conclusions of analyses done at different spatial scales remain consistent. These case studies demonstrate the utility of a Europe-wide weed survey database, for clarifying or extending results obtained from studies at smaller scales. This Europe-wide data collection offers many more opportunities for analysis that could not be addressed in smaller datasets; including questions about the effects of climate change, macro-ecological and biogeographical issues related to weed diversity as well as the dominance or rarity of specific weeds in Europe.

Questions: Two scientific disciplines, vegetation science and weed science, study arable weed vegetation, which has seen a strong diversity decrease in Europe over the last decades. We compared two collections of plot-based vegetation records originating from these two disciplines. The aim was to check the suitability of the collections for joint analysis and for addressing research questions from the opposing domains. We asked: are these collections complementary? If so, how can they be used for joint analysis?. Location: Europe. Methods: We compared 13 311 phytosociological relevés and 13 328 records from weed science, concerning both data collection properties and the recorded species richness. To deal with bias in the data, we also analysed different subsets (i.e., crops, geographical regions, organic vs conventional fields, center vs edge plots). Results: Records from vegetation science have an average species number of 19.0 ± 10.4. Metadata on survey methodology or agronomic practices are rare in this collection. Records from weed science have an average species number of 8.5 ± 6.4. They are accompanied by extensive methodological information. Vegetation science records and the weed science records taken at field edges or from organic fields have similar species numbers. The collections cover different parts of Europe but the results are consistent in six geographical subsets and the overall data set. The difference in species numbers may be caused by differences in methodology between the disciplines, i.e., plot positioning within fields, plot sizes, or survey timing. Conclusion: This comparison of arable weed data that were originally sampled with a different purpose represents a new effort in connecting research between vegetation scientists and weed scientists. Both collections show different aspects of weed vegetation, which means the joint use of the data is valuable as it can contribute to a more complete picture of weed species diversity in European arable landscapes.

Long-term national European weed surveys, large scale classical phytosociological programs and camera-based documentation systems lead to results which can be documented in form of maps. Comparisons of these visual representations of relative weed positions can be used for the prediction of changing weed spectra and of plant biodiversity changes. Statistical methods connected with mapping software are used for the analysis of environmental factors and of farm managing practices influencing the occurrence of weeds. Maps produced by sensor-driven weed detection devices still differ considerably from maps produced via classical phytosociological approaches. Computer algorithms may allow the precise identification of some weeds in camera images. The present technical solutions are, however, still far from those achieved by experienced botanists. Many weed detection tools based on algorithms are not able to distinguish between closely related weeds yet. A few European countries have a long tradition of surveying weeds in major crops by traditional tools. Various software packages are employed for the analysis, documentation and visualisation of survey results. Large scale comprehensive maps including the infestation of crops over different countries are, however, often biased as not every national research group uses the same methods for the assessment of weed infestation. The ranking of the most common species seems, however, to allow comparable conclusions. The recognition of trends in spectrum changes can only be derived from long term studies as we see it. Our review reflects discussions within the Weed Mapping Working Group of the European Weed Research Society over the last ten years. We try to identify new research trends and to respond accordingly with new research projects. What we see today is a shift from traditional mapping approaches towards the use of digital devices as for example in precision farming projects. Another issue of increasing importance is the mapping of herbicide resistant biotypes.

“Arable Weeds and Management in Europe” is a collection of weed vegetation records from arable fields in Europe, initiated within the Working Group Weeds and Biodiversity of the European Weed Research Society (EWRS). Vegetation-plot data from this scientific community was not previously contributed to databases. We aim to prove the usefulness of collection for large scale studies through some first analyses. We hope to assure other weed scientists who have signalled willingness to share data, and plan to construct a full data base, making the data available for easy sharing. Presently, the collection has over 60,000 records, taken between 1996 and 2015. Many more studies for potential inclusion exist. Data originate mostly from studies exploring the effect of agricultural management on weed vegetation. The database is accompanied with extensive meta-data on crop and weed management on the surveyed fields. The criteria for inclusion were a minimum amount of information on the cultivated crop, and a georeference. Most fields were surveyed repeatedly, i.e. transects, multiple random plots, or repeated visits. All surveys aimed to record the complete vegetation on the plots. Sometimes, taxa were identified only to genus level, due to survey dates very early in the vegetation period. Plant taxonomy is standardized to the Euro+Med PlantBase.

To better manage invasive populations, it is vital to understand the environmental drivers underlying spatial variation in demographic performance of invasive individuals and populations. The invasive common ragweed, Ambrosia artemisiifolia, has severe adverse effects on agriculture and human health, due to its vast production of seeds and allergenic pollen. Here, we identify the scale and nature of environmental factors driving individual performance of A. artemisiifolia, and assess their relative importance. We studied 39 populations across the European continent, covering different climatic and habitat conditions. We found that plant size is the most important determinant in variation of per-capita seed and pollen production. Using plant volume as a measure of individual performance, we found that the local environment (i.e. the site) is far more influential for plant volume (explaining 25% of all spatial variation) than geographic position (regional level; 8%) or the neighbouring vegetation (at the plot level; 4%). An overall model including environmental factors at all scales performed better (27%), including the weather (bigger plants in warm and wet conditions), soil type (smaller plants on soils with more sand), and highlighting the negative effects of altitude, neighbouring vegetation and bare soil. Pollen and seed densities varied more than 200-fold between sites, with highest estimates in Croatia, Romania and Hungary. Pollen densities were highest on arable fields, while highest seed densities were found along infrastructure, both significantly higher than on ruderal sites. We discuss implications of these findings for the spatial scale of management interventions against A. artemisiifolia.