Among the vast array of plant species within the Ambrosia genus, Ambrosia artemisiifolia L. emerges as a particularly troublesome species (Goeden and Andrès, 1999; Taylor, 2019). Originating in North America, this invasive plant from the Asteraceae family has rapidly expanded its reach, becoming a ubiquitous presence in diverse ecosystems around the world (Smith et al., 2013; Montagnani et al., 2017). Characterized by its robust vigor, superior adaptability, high seed yields, and rapid dispersal, A. artemisiifolia presents an important menace capable of significantly altering the dynamics of native flora and fauna (Essl et al., 2015). Over the last two centuries, A. artemisiifolia has spread in 80 countries, disrupting ecological balances, diminishing biodiversity, and threatening the production of valuable farmland, thereby reducing crop yields (Dong et al., 2020). The adaptability potential of A. artemisiifolia takes various forms, including genetic diversity and phenological variations. Studies have consistently highlighted the high genetic variability of this species, pointing to its capacity to adapt to diverse environments (Genton et al., 2005; Li et al., 2019; Sun et al., 2020). Beyond genetic diversity, also the phenological variations are further accentuating the species’ ability to thrive across varied ecological niches (Buttenschøn et al., 2009; Leskovšek et al., 2012; Scalone et al., 2016). This adaptability assumes even greater significance considering predictions of A. artemisiifolia’s continued expansion fueled by global warming (Cunze et al., 2013; Skálová et al., 2017; Sun et al., 2020). In addition to its disruptive impact on natural ecosystems, A. artemisiifolia has proven to be a significant agricultural weed. Changes in agricultural practices, especially the shift towards more intensive farming systems, have provided a suitable environment for the accelerated expansion of this invasive species (Kiss and Béres, 2006). Studies indicate that high populations of A. artemisiifolia can result in substantial yield losses, reaching up to 75% in soybean and 80% in maize, compared to estimated weed-free yields (Weaver, 2001; Cowbrough et al., 2003). Various control measures including chemical, physical and biological techniques have been explored to mitigate the impact of A. artemisiifolia (Buttenschøn et al., 2009; Milakovic et al., 2014; Milakovic and Karrer, 2016). However, eradication of this species has proven to be a formidable challenge. Notably, A. artemisiifolia exhibits resistance to air and soil pollution (Pichtel et al., 2000; Ziska, 2002), and in some cases it remarkably retains the ability to reproduce even after damage from defoliation and mowing (Brandes and Nitzsche, 2006; Gard et al., 2013). Presently, control of this invasive weed species is feasible in major crops through chemical, mechanical, or combined measures (Scruggs et al., 2019; Máčajová et al., 2022; Wang et al., 2022). However, there is a concern over the potential development of herbicide resistance over time (Saint-Louis et al., 2005). Understanding the distribution and spread of this species is challenging. Its introduction to Europe was tied to trade, resulting in different introduction points at different times and contributing to genetic variations. Populations already present in Europe further propagated in a similar manner, fostering hybridization and population variations (Csontos et al., 2010; Chapman et al., 2016; van Boheemen et al., 2017; Afonin et al., 2018). Despite existing studies on the introduction and distribution of A. artemisiifolia in the origin areas of the studied populations (Chauvel et al., 2006; Csontos et al., 2010; Galzina et al., 2010; Gentili et al., 2017), pinpointing the exact introduction pathway and population evolution remains a difficult, if not impossible, task. Given the species’ great adaptability potential and diversification (Battlay et al., 2023), it might be more meaningful to explore the responses of different populations to climatic conditions.

To address this critical issue and improve the control of A. artemisiifolia, an in-depth study of its behavior and physiology is essential, particularly during crucial stages such as germination. Germination and subsequently the emergence are the most critical periods in weed species life cycle, as they can largely influence the future competition with the crops and damage to the yield (Forcella et al., 2000). These critical periods mostly depend on the temperature and water availability needed for germination processes. In this regard it was discovered that seeds have temperature and water thresholds below which germination is not possible, these thresholds are defined as base temperature (T_b) and base water potential (Ψ_b) respectfully (Arnold, 1959; Dahal and Bradford, 1994). This knowledge can inform the development of emergence models, aiding in predicting growth dynamics and suggesting optimal timing for control operations, thereby reducing herbicide use and airborne pollen levels (Masin et al., 2005, 2010, 2014; Colbach et al., 2007). Given the high variability of A. artemisiifolia, that reflects in both a wide range of temperatures and water availability needed for germination (Leiblein and Lösch, 2011; Dinelli et al., 2013; Knolmajer et al., 2024), it is crucial to determine if populations from different regions exhibit similar behavior. Considering that there are only a few articles found in the literature that examined these parameters (Guillemin et al., 2013), in this study, we compared four A. artemisiifolia populations, examining seed parameters and determining base temperature (T_b) and base water potential (Ψ_b) essential for germination, as well as different seed physical properties. By discovering the germination dynamics of diverse populations, our research aims to contribute to a broader understanding of A. artemisiifolia behavior. Insights gained can potentially inform the development of effective control strategies that are appropriate to specific populations, ultimately mitigating the ecological, agricultural, and health impacts of this invasive species.

Base temperatures and base water potentials of the studied populations of A. artemisiifolia are shown in Table 1 .

The analysis showed distinct trends in T_b across the tested populations ( Table 1 ). Notably, the Italian population exhibited the lowest T_b and was significantly different from the Serbian and French populations. Differently, the French population displayed the highest T_b, but was not statistically different from the Serbian and Croatian populations. The Croatian population had noteworthy variability, as evident from the standard error values.

The base water potentials (Ψ_b) among A. artemisiifolia populations exhibited more pronounced variations than the observed differences in T_b ( Table 1 ). The Ψ_b of the Italian and Croatian populations were the highest, with no significant difference between them. Conversely, the Serbian and French populations displayed considerably lower Ψ_b values, with the lowest being the one of the Serbian population.

The results of the one-way ANOVA on the physical properties are presented in Table 2 .

There were significant differences between the populations for every physical property tested ( Figure 3 ). The seeds belonging to the French population were the largest, followed by the Croatian population and then by the Serbian and Italian, with seeds of a similar size. It can also be seen that the results concerning seed perimeter confirm the size comparison. However, although similar in area, it can be observed that the perimeter is different between the seeds of Serbian and Italian population of this weed species.

Seed circularity is an insightful parameter offering information about seed shape. A value closer to 1 indicates a shape more akin to a perfect circle, while a value closer to 0 signifies a more elongated form. It is noteworthy that none of the tested populations exhibits perfectly round seeds. Among them, the Italian population comes closest to a circular shape, while the French population showcases the most elongated seeds.

Finally, the results show a distinctive seed weight for each population. Surprisingly, the seeds of the Croatian population, despite not being the largest, emerge as the heaviest among the populations. In contrast, the seeds of the French population, which are the largest in size, rank third in terms of weight. Remarkably, the lightest seeds are from the Italian population, aligning with the smaller size characteristic of this population’s seeds.

The results revealed disparities in both germination thresholds and physical properties among various populations of A. artemisiifolia. This aligns with prior studies highlighting the weed species considerable genetic and phenotypic variability (Leiblein-Wild and Tackenberg, 2014; Martin et al., 2016; Li et al., 2019). Rapid adaptation to new environments is a common trait among invasive weed species (Prentis et al., 2008; Clements and Jones, 2021; Gong et al., 2022). The introduction of A. artemisiifolia to Europe from its native habitat employed an adaptation strategy, establishing it as a formidable invasive species globally. Studies on A. artemisiifolia adaptation underscore its exceptionally high potential, with adaptations occurring rapidly, contributing significantly to its invasiveness (Battlay et al., 2023). These findings are in accordance with the results obtained in this study, revealing interesting differences and similarities among the studied populations.

The influence of climate and climate change on the behavior of A. artemisiifolia is well-documented (Hamaoui-Laguel et al., 2015; Case and Stinson, 2018; Gentili et al., 2019; Xian et al., 2023), consistent with the results obtained in this study. Analyzing meteorological data on minimum, maximum, and mean temperatures in the four areas ( Figure 2 ) reveals that Turin and Zagreb experience similar temperatures. Conversely, Šabac tends to have higher temperatures in the spring-autumn period, while Dijon tends to have lower temperatures during the same period compared to Turin and Zagreb. This may explain the similarities in T_b for the Turin and Zagreb populations, although the Croatian population shows significant variability, indicating a more diverse response. Notably, the spring temperatures for Zagreb intersect with those of other places, potentially explaining its wide germination range.

For the Serbian and French populations, Šabac experiences very low winter temperatures that rise rapidly in spring, remaining high until autumn. While the temperatures in Dijon are consistently lower than in other areas, the consistently high minimum temperatures may explain why this population has the highest T_b. This is in accordance with different studies conducted on this matter, explaining how T_b is closely linked with the temperature of the habitat, indicating that the plants, even of the same species growing under different temperature regimes will have different T_b, ergo plants growing in colder climates will have lower T_b compared with those growing in warmer climates (Trudgill et al., 2000; Bao et al., 2019). This finding highlights important considerations for effective weed control strategies. For instance, in Dijon, the combination of consistently high and steady minimum temperatures during the early months corresponds to a population with a higher T_b. This likely delays seedling emergence compared to populations from warmer regions, as emergence depends on the accumulation of sufficient thermal time above T_b (Grundy, 2003; Baraibar et al., 2018). Such delayed emergence may help A. artemisiifolia in Dijon avoid frost damage or other unfavorable early-season conditions, allowing optimal emergence and development during warmer months. This pattern aligns with adaptive strategies observed in other weed species (Ramesh et al., 2017). However, this delayed emergence has practical implications for weed management, necessitating adjustments in the timing of control measures to coincide with peak seedling emergence. Emergence predictive models can play a crucial role in this regard. In contrast, populations with lower T_b might emerge earlier, leading to differences in developmental stages across regions, further emphasizing the need for region-specific management strategies. Examining the Ψ_b of the studied species and observing precipitation data in Figure 2 , it is evident that Turin receives the most precipitation in April and May, aligning with June precipitation in Zagreb. This explains the high Ψ_b of the Italian population and its similarity to the Croatian one. In contrast, Šabac receives significantly less precipitation during the same period, establishing it as the driest area compared to others. This likely accounts for the low Ψ_b of the Serbian A. artemisiifolia population, indicating adaptation to drought conditions. As for T_b, similar studies were conducted on the relationship between the Ψ_b of plants and water availability in different habitats. These studies also concluded that the seeds developing in more arid conditions will have a more negative Ψ_b compared to those, even of the same species, growing in more humid conditions (Ceri and Etherington, 1991; Baskin and Baskin, 2014a; Bao et al., 2022). Interestingly, despite high precipitation in the Dijon area, its Ψ_b is the second lowest. This discrepancy may be attributed to lower temperatures in this region compared to others, emphasizing the crucial interaction between temperature and precipitation for germination thresholds. A similar finding can also be observed in the work by Dürr et al. (2015), who observed that species growing in cool conditions were more able to germinate in dry conditions than those growing in warmer conditions.

It is important to emphasize that the seeds used in these trials were the original seeds collected from their native sites and were not reproduced in a common garden or uniform environment. This approach was chosen to capture the seeds local responses, reflecting the base temperature (T_b) and base water potential (Ψ_b) specific to their place of origin. By including both external (climatic) and internal (genetic/maternal) effects influencing these parameters, this work aimed to provide results that are representative of real-world conditions. This is particularly critical given that emergence predictive models are designed for specific areas and require calibration to ensure their accuracy and efficiency in those environments (Masin et al., 2010, 2012, 2014; Šoštarčić et al., 2021b).

The results regarding the physical properties of the seeds are consistent with studies on invasive weed species, demonstrating high variability among different populations (Fenollosa et al., 2021). These findings parallel the results of germination thresholds and can be interpreted as an adaptation to different climatic conditions. Qiu et al. (2010) found a positive correlation between high precipitation and low temperatures, influencing seed size. This aligns with our results, as French seeds turned out to be the largest. The same authors also noted that larger seeds typically have lower temperature requirements for germination, contradicting our findings. However, our results align with those of Ge et al. (2020), who discovered that some larger seeds might have higher thermal requirements for germination.

Several studies investigating the influence of the origin of A. artemisiifolia have highlighted temperature as a major driver of seed variability between different populations. Typically, seeds from higher latitudes and lower temperatures produce larger seeds with lower germination potential compared to populations from lower latitudes and higher temperatures (Li et al., 2015; Zhou et al., 2021), consistent with the results obtained in this study.

In discussing the generalizability of our findings across different populations of A. artemisiifolia, it is important to consider the reasons behind the selection of the four populations used in this study. Previous research has shown that populations of A. artemisiifolia exhibit very low variability when situated within the same geographic niche, even when separated by several kilometers (Scalone et al., 2016; Song et al., 2023; Putra et al., 2024). A similar niche effect was also found by Farooq et al. (2019), according to the authors, this effect is a consequence of ecological adaptations to prevailing climatic conditions in the area. This consistency in population characteristics supports our decision to focus on populations from distinctly different regions across Europe. Additionally, the niche-filling tendencies of this species imply that significant variations are more likely to arise from populations that are very far apart or exposed to very different conditions, differences and similarities between the populations uncovered in this work are clear evidence in support of this claim. This context is crucial as it not only underpins the reliability of our data, suggesting that the observed differences in germination thresholds and seed properties are likely attributable to regional climatic conditions rather than mere geographic proximity. But it also highlights one of the main limitations of emergence predictive models—‘one model cannot fit all’—emphasizing the need to develop specific models tailored to different areas.

The results presented here broaden our understanding of the behavior of this important invasive species, but can also be considered as an important contribution to effective control strategies. As it was mentioned previously, different control strategies for A. artemisiifolia are present, including chemical, mechanical and biological methods, that can also be combined, with varying degrees of effectiveness (Buttenschøn et al., 2009; Gerber et al., 2011; Essl et al., 2015). Considering that the chemical weed control methods are often the most effective ones, it is no surprise that they are the most used ones, also for the control of this species. However, the need to control A. artemisiifolia also outside of the strictly agricultural context, both effective and environmentally friendly control becomes particularly challenging (Essl et al., 2015; Wang et al., 2022).

Using the temperature and water potential thresholds determined in this work, it could be possible to create emergence prediction models for the tested areas, to assist the farmers and authorities in choosing the best timing for control operations, thus maximizing the effectiveness and reducing the excessive use of herbicides (Masin et al., 2010, 2012). Moreover, the same principle could also be applied for mechanical control methods and integrated weed management (IWM) strategies, providing the species emergence percentage in time. An example of the usefulness of choosing the right time for weed control operations can be observed in the work of Beam et al. (2021), who found that the effectiveness of A. artemisiifolia control using IWM strategies strongly depends on the time of their application. Having in mind that today the timing of control operations against this weed species is decided arbitrarily, an emergence predictive model based on the data obtained in this study could significantly improve this weed management strategy.

This study sheds light on the germination thresholds and diversity among different European populations of A. artemisiifolia, offering crucial insights into the variability of this invasive species. The results demonstrated significant differences in germination thresholds, with the Italian population exhibiting the lowest base temperature (0.58°C), while the French population demonstrated the highest (2.74°C), and the Serbian population showing the lowest base water potential (−1.44 MPa), while the Italian one exhibited the highest (−0.78). Additionally, substantial disparities were also observed in seed physical properties, where the Italian population had the smallest and lightest seeds, while the French population had the largest and most elongated seeds. Given its high adaptability and the associated threats to human and animal health, as well as agricultural production, the findings presented in this research carry substantial importance.

By addressing the gap in knowledge regarding the differentiation of A. artemisiifolia at a European level, considering diverse climatic conditions and geographic origins, this study contributes valuable information for the effective management of this weed species. The germination thresholds obtained can serve as foundational data for constructing emergence predictive models or enhancing existing ones, which are vital for timely control measures against this agricultural and health hazard in Europe.

Moreover, understanding the levels of seed variability provides a basis for refining control strategies and mitigating the spread of this invasive species. While this study offers valuable insights, further research is needed, including more populations from various European countries. Future studies, particularly those simulating the effects of climate change, will contribute to a more comprehensive understanding of A. artemisiifolia’s adaptability. Such insights are crucial for anticipating the species’ impact on agricultural production and the overall well-being of European communities.