WEED COMPOSITION IN CONVENTIONALLY AND ORGANICALLY GROWN MEDICAL AND AROMATIC PLANTS

This study aimed to compare weed flora in conventionally and organically grown medicinal and aromatic plants (MAPs): basil, pot marigold, dill, and peppermint; in terms of weed composition and weed abundance. A total of 28 weed species (25 and 15 species in conventional and organic crops, respectively) were identified. The presence and abundance of certain weed species were affected by MAP species and farming system. Higher weed diversity and weediness, and lower floristic similarity were found in conventionally grown crops. Also, the analysed MAPs differed in weediness by individual weed species. Correspondence analysis pointed to conventional and organic MAPs with the most frequent and most abundant weed species. Setaria pumila and Portulaca oleracea were the most frequent species in conventional; and Amaranthus retroflexus , Datura stramonium , and Sorghum halepense in organic crops . In both conventional and organic farming systems, therophytes were the most dominant life forms indicating a strong anthropogenic influence. The results should contribute to establishing weed control measures that are adequate for the two farming systems.


INTRODUCTION
. Weed composition in conventionally and organically grown medical and aromatic plants. Acta Sci. Pol. Hortorum Cultus, 21(4), 115-126. https://doi.org/10.24326/asphc.2022.4.12 MAPs [Carrubba and Catalano 2009, Carrubba and Militello 2013, Hendawy et al. 2019. Chemical weed control is limited in MAPs, even in conventional farming; therefore, the production has increasingly been based on the principles of organic farming, which prohibits the use of chemicals according to EU regulations [EEC 1991, EEC 1992. Even though there are clear guidelines for organic crops management, it is often difficult to maintain weed populations below the tolerance threshold. Therefore weed control in organic farming is based on direct and/or indirect alternative non-chemical measures: agronomic, biological, and physical [Barberi 2002, Carrubba andMilitello 2013].
Weed composition depends on several factors: crop species, weed control measures, environmental conditions, soil fertility, climate, etc. [Roschewitz et al. 2005, Barroso et al. 2015, Pinke et al. 2016, Travlos et al. 2018, Hendawy et al. 2019. Understanding how the farming system affects weed composition is a necessary step for transfer from reactive to predictive weed control management [Barroso et al. 2015]. Also, climate changes may cause changes in the composition of weeds, e.g. certain species may become invasive [Scott et al. 2014]. Because of the dynamic nature of crop weediness, constant monitoring gains in importance. Identification of dominant weeds in MAPs and understanding their biology and ecology is a necessary step forward to the establishment of the appropriate management strategies [Hendawy et al. 2019].
This study aimed to compare weed flora in conventionally and organically grown MAPs (basil, pot marigold, dill, and peppermint), in terms of weed composition (species richness, floristic diversity, evenness, dominance) and weed abundance (weediness, frequency). The results should contribute to establishing appropriate weed control measures.

MATERIAL AND METHODS
Floristic study of weeds in medicinal and aromatic plants (MAPs): basil (Ocimum basilicum), pot marigold (Calendula officinalis), dill (Anethum graveolens), and peppermint (Mentha × piperita), grown in conventional and organic farming was carried out in three growing seasons (2014,2015,2016), at the experimental fields of the Institute of Field and Veg-etable Crops, National Institute of the Republic of Serbia, Alternative Crops and Organic Production Department, in Bački Petrovac (Vojvodina province,Serbia;45.36° N,19.62° E; elevation 86 m; continental climate). The study was set up in two experimental fields, conventional and organic. Each plant species was grown in both production systems. The main plots were of the same size for all crops and consisted of 12 rows, 25 m long, with the between-row spacing of 70 cm. The main plots were replicated three times for each variant. The organic field of 7 ha was established in 2009, following a two-year conversion period that took place in compliance with all the principles of organic agriculture. The practice included organic fertilization regime, integrated non-chemical weed and pest management, crop rotation, intercropping, floral corridors; all adapted to the specific cultivated plant species. The organic field was under crop rotation regime which included legumes, small grains, and various MAPs in consecutive years. The preceding crop was winter barley in all three seasons. In the conventional plot, the MAPs were rotated randomly every year. The organic field was fertilized with 15 t ha -1 farmyard manure applied in November 2011, and the conventional plot was fertilized every autumn with 400 kg ha -1 NPK 15 : 15 : 15. The fields were ploughed and harrowed at the end of every season. Weather data and the results of the soil analyses are given in Figure 1 and Table 1, respectively. The weeds were not controlled during the study.
Species names are given according to Nikolić [2015]. The status of weed invasiveness for Europe and Serbia, as well as the invasive species of global importance, were given according to DAISIE [2018], Lazarević et al. [2012], and GISD [2019], respectively. Taxonomic affiliation of species to families, life forms, and time of flowering are given according to Takhtajan [2009], Ujvárosi [1973], and Landolt et al. [2010], respectively.
where are:   (1) (3) (8) Ljevnaić-Mašić, B., Brdar-Jokanović, M., Džigurski, D., . Weed composition in conventionally and organically grown medical and aromatic plants. Acta Sci. Pol. Hortorum Cultus, 21(4), 115-126. https://doi.org/10. 24326/asphc.2022.4.12 N -the total number of individuals of all species in the plot; Z i -number of sampling plots with species i present; Z -total number of sampling plots surveyed; p i -a proportional abundance of i th species; ln -natural logarithm; S -species richness (total number of species); a -total number of species in the first population; b -total number of species in the second population; c -number of common species; A -total number of individuals in population A; B -total number of individuals in population B; W -sum of the lower of the two abundances of each species in the population.
Differences in weediness between conventionally and organically grown MAPs and their interactions were determined using ANOVA and t-test. Correspondence analysis was used to determine the relationship between the presence and abundance of weed species in the studied MAPs. All calculations were performed using Statistica 13.2 software [Dell™ Statistica™ 13.2 University License].

Weed composition.
Floristic study of weeds in basil, pot marigold, dill, and peppermint cultivated in conventional and organic production systems, found a total of 28 species, 25 species of which in conventional and 15 species in organic farming (Tab. 2). All weeds belong to 15 families. All of them are invasive species in Europe, while Ambrosia artemisiifolia, Cirsium arvense, Senecio vulgaris, and Sorghum halepense are invasive species of global importance. Five species are invasive in Serbia: very invasive Ambrosia artemisiifolia, sporadically invasive Sorghum halepense, and species Amaranthus retroflexus, Datura stramonium, and Veronica persica which are, currently classified as potentially invasive in Serbia.
Conventional basil (16 species) and conventional peppermint (15 species) were with the highest species richness, while conventional and organic pot marigold and organic basil were with the lowest number of weed species (7 species each). The therophytes were the most dominant life forms in all studied crops (70-90%). Weediness was significantly higher in conventional than in organic crops. Conventional crops were mostly weeded by Setaria pumila and Portulaca oleracea, and organic crops by Amaranthus retroflexus. The most frequent weed species in both MAP farming systems were Amaranthus retroflexus, Chenopodium album, and Senecio vulgaris with a frequency of 100%.
The floristic diversity of weeds was higher in conventional MAPs (H' = 2.46), with higher evenness (E = 0.76) and lower dominance of the most frequent weed species (D = 0.12) than in organic crops (Tab. 2). The weeds in the organic crops were floristically more similar than in the conventional crops (Fig. 2). When comparing the weeds composition relative to the farming systems the highest floristic similarity was found for conventional basil and pot marigold (S J = 43.75%), and organic basil and peppermint (S J = 70.00%), Figure 2a. When comparing the composition of weeds relative to MAPs, pot marigold crops were most similar (S J = 75.00%), while dill crops were least similar (S J = 19.05%). On the other hand, Steinhaus's coefficient index (S A ), which takes into account both the total number of individual plants and abundance of species, showed that organic dill and peppermint (S A = 0.6277) were most similar regarding weed composition and abundance (Fig. 2b).
Weed flowering time is significant for a timely application of adequate weed control measures. The majority of identified weeds start flowering in June (11 weeds) and July (7 weeds), mostly ending flowering in September (16 weeds) and October (10 weeds), Figure 3. The earliest flowering with the longest flowering period (February to October) has Veronica persica, while Polygonum aviculare has the shortest flowering period (May to July). Ambrosia artemisiifolia starts flowering the last (in August) and ends in October.
Statistical analyses. ANOVA and t-test determined significant differences in the weediness among all the analyzed variables: weed species, MAPs, and farming systems. The interactions were also significant (Tab. 3, Fig. 4, Fig. 5). Weediness, i.e. the frequency and abundance of certain weed species in MAPs, was affected by crop species and farming system, which was confirmed by correspondence analysis (Tab. 2, Fig. 6). The correspondence analysis separated conventional and organic MAPs with the most frequent and most abundant weed species. Conventional dill differed  from other MAPs, probably due to its highest floristic diversity (H' = 2.34) and the highest evenness of weed flora (E = 0.91), without the domination of the most abundant weed species (Tab. 2, Fig. 6).
Higher floristic diversity most often causes higher evenness, however lower dominance of the most abundant species, and vice versa [He and Legendre 2002, Magurran 2004, Jost 2010]. This was confirmed in this study. All evenness values were relatively high and did not differ much, which resulted in low dominance values of the most abundant species in both farming systems. Floristic similarities found in MAPs were the result of specific and similar conditions of their cultivation. However, certain authors report species richness to be negatively correlated with evenness [Stirling and Wilsey 2001, Magurran 2004, Zhang et al. 2012. Therefore, according to He and Legendre [2002], research into the mechanisms that affect evenness can facilitate understanding the species diversity.
In conventional farming systems, certain weeds are ecologically well-adapted and able to survive agronomic selective pressures, and therefore their control is more difficult. According to Nkoa et al. [2015], lower floristic diversity within an agroecosystem can lead to its greater vulnerability. The situation allows new species to inhabit, and this was confirmed in our study. Invasive species Sorghum halepense, as well as Amaranthus retroflexus, and Datura stramonium, potentially invasive weed species in Serbia, were among the most abundant weeds in organic MAPs which had lower floristic diversity. Possible reasons for the greater presence of invasive weeds in organic MAPs may be apparently still unbalanced agroecological conditions found in these recently established organic plots, and crop rotation [Brdar-Jokanović et al. 2018]. Although preceding crop partially suppressed weed growth in organic plots, invasive species in organic MAPs developed first and spread fast due to their extreme adaptability. Additionally, the organic field remained less fertile than the conventional plot, despite the long-term beneficial effects of farmyard manure on soil properties. Requirements of weeds for nutrients are species dependant [Travlos et al. 2018]. For example, the competitive ability of Amaranthus retroflexus increases with higher levels of nitrogen [Blackshaw and Brandt 2008], which is consistent with our study where Amaranthus retroflexus was one of the most frequent and most abundant weeds in both farming systems. Hence, low nitrogen fertilization could decrease the density of Amaranthus retroflexus and enhance the effectiveness of weed control [Travlos et al. 2018].
The majority of MAPs are less competitive to weeds, which causes substantial weediness with annual species -therophytes [José-María et al. 2011]. In this study, the therophytes were the most common life forms in all studied crops. The dominance of therophytes points to intensive anthropogenic effects, i.e. intensive cultivation practices in both farming systems, which has led to their good adaptability. The adaptability is reflected by longer flowering period, a tendency for seed dispersal, and low harvest index [Carrubba andMilitello 2013, Abouziena andHaggag Wafaa 2016]. Other authors found that annual species are associated with a range of tillage systems, while perennial species are associated with reduced-and zero-tillage systems , Thomas et al. 2017, Travlos et al. 2018].
There is no single method that could be used to provide highly efficient weed control under the organic farming system. Due to the specific cultivation of MAPs, the following control measures are recommended: pre-sowing irrigation, sowing methods, hilling, hand weeding, drip irrigation method, natural herbicides, hot water, soil mulch, etc. [Carrubba andMilitello 2013, Hendawy et al. 2019]. Further, there is a need for alternative weed management depending upon the weed composition, farming system, and management practices such as fertilization and tillage [Baker et al. 2018, Travlos et al. 2018]. Since the recorded weeds flower, until the development of more efficient methods that are in compliance with the principles of organic farming, mechanical weeding should be performed at least three times during the growing season [Brdar-Jokanović et al. 2018].

CONCLUSION
This is the first comparative analysis of weed composition and abundance between conventional and or- . Weed composition in conventionally and organically grown medical and aromatic plants. Acta Sci. Pol. Hortorum Cultus, 21(4), 115-126. https://doi.org/10. 24326/asphc.2022.4.12 ganic farming systems of MAPs (basil, pot marigold, dill, and peppermint) in agro-ecological conditions of Serbian province Vojvodina. It represents a step towards establishing adequate weed control measures. Weed composition and abundance were affected by MAP species and farming system. The crops under conventional production systems were characterized by higher weed diversity and weediness, and lower floristic similarity, comparing to organic crops. Weediness by individual plant species differed among the studied MAPs. The most frequent species in conventional crops were Setaria pumila and Portulaca oleracea. Amaranthus retroflexus, Datura stramonium, and Sorghum halepense were the most frequent species in organic crops. Therophytes were the most dominant weed life forms.
The results of this study could be helpful for the assessment of appropriate weed control measures. Further studies are required to understand the weed effects on MAP yields and quality; as well as to confirm the hypothesis on the suppressing effects of certain MAPs. Furthermore, it is necessary to raise awareness on the importance of conserving weeds biodiversity as an integral part of balanced agroecosystems.