Differential Germination and Growth Response to Temperature of Three Ambrosia Weed Species—Implications for Future Spread

Three main Ambrosia species (Ragweed) grow in Israel; the most abundant invasive Ambrosia confertiflora DC, whereas A. artemisiifolia L. and A. tenuifolia Spreng., are of restricted distribution. The present research was aimed to study the effect of temperatures regimes on the development and growth of these Ambrosia species, to elucidate the environmental conditions and plant traits that affect their growth and infestation patterns. All three Ambrosia species germinate best in light from the soil surface with no prerequisite of a stratification period. A. confertiflora seed emergence is inhibited at high temperature regimes (28/34°C). A. artemisiifolia at low temperature regimes (10/16°C), while A. tenuifolia is less affected by the temperature regimes. A. confertiflora plant height increases with increasing temperatures, and at lower temperatures develops a rosette. Root and rhizome biomass were less affected by the different temperatures regimes; A. artemisiifolia aboveground mass was not affected by temperature regimes while A. tenuifolia aboveground mass was reduced only at lower temperatures. A. confertiflora fast invasion and establishment are due to the combined effects of prolific seed dispersal, rapid sprouting and growth, and its phenotypic plasticity.


INTRODUCTION
Biological invasions are a main threat to biodiversity in terrestrial, freshwater, and marine ecosystems, requiring extensive research and management efforts (Mačić et al., 2018). Invasive species are defined as animals, plants, or other organisms introduced by man into places out of their natural range of distribution, where they become established and disperse, generating a negative impact on the local ecosystem and species. A. artemisiifolia is known throughout the world as an invasive weed and in many countries it causes severe damages to agriculture and substantial health effects to millions of people, due to its high allergenic properties (Makra et al., 2005(Makra et al., , 2014Bullock et al., 2012;Montagnani et al., 2017;Schaffner et al., 2020). In Israel there are five Ambrosia species, but only A. confertiflora is defined as an invasive species (Dufour-Dror, 2016;Yair et al., 2019).
The origin of most Ambrosia species is in the Americas and they spread mainly by global commerce to Europe, Asia and Australia. The genus Ambrosia belongs to the Asteraceae family (Compositae) and includes 42 species (Strother, 2006). Ambrosia plants are annual or perennial monoecious with wind-pollinated flowers producing achenes. The wind-borne pollen grains cause severe allergenic response in humans and animals. The perennial species propagate by achenes and by underground rhizomes.
There are five Ambrosia species in Israel, which were detected in the early 1990's (Danin, 1994;Yair et al., 2017). Species details and characteristics are summarized in Table 1. The invasive perennial species A. confertiflora DC (Burr ragweed), is the most abundant in Israel, mainly found in Central Israel; Three naturalized perennial species A. tenuifolia Spreng; (Lacy ragweed) is found in several locations but its spread is limited; A. psilotachya DC (Cuman ragweed or Perennial ragweed and A. greyi (wooly leaf bur ragweed) are restricted to one location only. The fifth species A. artemisiifolia L. (short ragweed, common ragweed) an annual casual plant that appears in several sites but manages to survive only in the Hula Lake Natural Reserve in northern Israel. The causes of Ambrosia spp. invasions in Israel are mainly anthropogenic and are believed to originate with shipments of grain for food, feed and oil production from the USA. A, confertiflora is hardly studied. This perennial species originated in South-West US to Central Mexico and is invasive in Australia (Watt, 1987) and Israel (Yair et al., 2019), and may have a spread potential in south Europe (Tanner et al., 2017). A. confertiflora is a highly variable and complex erect plant, reaching up to 3 m in height, (Payne, 1964;Whisenant, 2018), which produces both achenes and rhizomes that create dense sprouts. The plant produces numerous viable hooked prickled achenes (1-2 mm long, as described in Yair et al., 2019) that are easily dispersed by animals and humans. A. tenuifolia produces rhizomes and achenes but the latter are nonprickled, hardly viable and thus their low dispersion potential (Yair et al., 2019).
Research regarding biological invasions and traits is mainly related to the A. artemisiifolia species that produces many achenes, 3,000-62,000 per plant (Dickerson and Sweet, 1971;Fumanal et al., 2007). Germination of seeds is optimal at soil level, decreases between 2 and 8 cm depth while in 10 cm depth no germination is observed, thus seed burial by deep tillage can reduce seed germination (Bassett and Crompton, 1975;Guillemin and Chauvel, 2011). Seeds of A. artemisiifolia require stratification at 4 • C for 8 weeks or more and for maximum germination there is a preference for alternating temperatures and light (Willemsen and Rice, 1972;Willemsen, 1975). The minimum temperature requirement for germination is 5 • C (Baskin and Baskin, 1980) and at optimal conditions, germination begins at 2 days (Ortmans, 2016). Seeds possess secondary dormancy (Bazzaz, 1970(Bazzaz, , 1979 and can germinate even after 40 years (Baskin and Baskin, 1977;Telewski and Zeevaart, 2002). The aim of this research was to study the effect of temperature on the development and growth of A. confertiflora compared to the other Ambrosia species present in the Israeli flora, to elucidate the environmental conditions and plant traits that may explain the rapid distribution and vast infestation of A. confertiflora in Israel.

Germination and Emergence Experiments
Plants and seeds of the three Ambrosia species A. confertiflora, A. artemisiifolia, and A. tenuifolia, were collected from various open field areas between the years 2009-2014 and kept dry in paper bags at 4 • C until use. For germination experiments, seeds were germinated in Petri dish on various substrates: Whatman No. 1 filter paper, sea sand, vermiculite, perlite, Rehovot sandy soil, and commercial potting soil (Tuff Merom Hagolan). Following the finding that the best substrate for germination was Whatman No. 1 filter paper, all experiments were conducted on two layers of this filter paper in 9 cm Petri dish, moistened with 500 microliter tap water, sealed with Parafilm and placed in a 25/18 • C 16d/8n growth chamber. No germination in Petri dishes wrapped with aluminum foil to prevent light penetration was recorded hence all other experiments were conducted under light condions. Seeds were determined germinated once the radicle length reached the length of the achene. For depth emergence experiments, plastic boxes (18 × 14 × 10 cm L/W/H) filled with sieved Rehovot sandy soil (1.2% clay, 3.3 % silt, 95.5% sand, 0.2% organic matter, pH 7.5) were used. For each box seeds of fifty A. confertiflora, or 20 A. artemisiifolia or A. tenuifolia were separately sown in rows at 0, 1, 2, 3, 4 cm soil depth, and placed in the phytotron under the following temperature regimes: 10/16 • C, 16/22 • C, 22/28 • C, and 28/34 • C night/day temperatures which represent the four seasons in the Mediterranean region. Pots were randomly placed, watered by hand with no fertilizer for 55 days. Emerging seedlings were recorded on a weekly basis.

Plant Development
For Ambrosia biomass experiments, four to six leaf A. confertiflora, A. artemisiifolia, and A. tenuifolia seedlings from FIGURE 1 | A. confertiflora emergence rate at four temperature regimes, 55 DAS. Columns within each temperature regime topped by different letters are significantly different according to Tukey-Kramer HSD, p = 0.05.

FIGURE 2 |
The effect of temperature regimes on the seedling emergence rate of A. confertiflora. Seed placed on the soil surface (0 cm seeding depth), 55 DAS. Three-parameter logistic nonlinear regressions between A. confertiflora emergence and time (days) were used. Parameters of the equations of four temperatures regimes are presented in Table 2.
Frontiers in Agronomy | www.frontiersin.org each species, were transplanted into 3.9 L plastic pots filled with potting soil and placed in the phytotron under four temperature regimes as described above. Eight pots (= 8 replicates) were transferred to each temperature regime. Pots were randomly placed and watered by hand for 121 days. Weekly measurements of plant height, sprout number, inflorescence development and final plant mass (shoot, root, and rhizomes) were conducted. The fresh weight of plants parts was determined at harvest and the dry weight after placing the plants in an oven for 3 days at a temperature of 60 • C. For convenience all results are presented in fresh weight.

Rhizome Development
To determine rhizome sprouts development of the two perennial species A. confertiflora and A. tenuifolia, 10 replicates of a single 6-8 leaf stage plant were transferred into the center of 50 × 30 × 6.5 cm plant trays containing potting soil, trays placed in the Faculty of Agriculture net-house in Rehovot, Israel. The Number of sprouts was recorded weekly until 111 DAS.
We conducted two experiments in the net-house at the same time of year-August to November in two subsequent years. The average of the two experiments is presented in the results.

Statistical Analysis
All experiments were conducted in a complete random design, repeated at least twice with five to six replications. The results were combined and analyzed as 10-12 replications. Analysis of variance was computed and means were compared using Tukey Kramer HSD test (P ≤ 0.05) using JMP ver. 13.0 (SAS Institute Inc. USA). When data were naturally quantative, a three-parameter nonlinear logistic equation was fitted using Sigma Plot 10 (SYStat Software, Inc. California, USA) (Ephrath et al., 2012): Where f (x) represents the number of Ambrosia spp. emergence, a represents the upper asymptote (maximum emergence), x 50 represents the time when Y is 50% of maximum, and b represents the slope at x 50.

Germination and Emergence Rates
In our Petri dish study, both A. artemisiifolia and A. confertiflora did not show a dormancy period thus do not require stratification. Attempts to enhance seed germination by physically cracking the seed coat, soaking in sodium   Table 3. hypochlorite, ethanol or gibberellic acid, were not successful. Seed germination initiated 2-3 days after seeding (DAS) and reached the maximum rate at 10 DAS. The average germination rate of A. confertiflora in Petri dish in light conditions was 40 and 56% for A. artemisiifolia. In dark conditions, A. confertiflora germination rate was only 0.4% and for A. artemisiifolia only 22%, indicating that both Ambrosia species require light for germination as was found at (Baskin and Baskin, 1980). The achenes of these species do not require stratification for germination and stratification did not increase germination rate. Achenes of the plant germinated immediately after maturing and harvest. Dry achenes maintained viability for a long period. Three out of four A. confertiflora achenes collected by A. Danin in Yamit, Sinai in 1981 and kept at the Jerusalem Herbarium, successfully germinated in our studies 35 years later and produced viable plants.
The emergence of A. confertiflora achenes (seeds) in the pot experiment started 6 days after seeding and the highest emergence rate of 30% was recorded in the 10/16 • C regime when the achenes were placed on the soil surface exposed to light. At this temperature regime, the burial of seeds at 1 cm depth was sufficient to drastically reduce seed emergence to 2% and at 4 cm depth, no emergence was recorded at all, indicating the importance of light for A. confertiflora emergence (Figure 1).
Seed emergence of A. confertiflora placed on the soil surface was inhibited at the highest temperature regime resulting in ca 5% emergence after 55 DAS, whereas at the lower temperature regimes the accumulated seed emergence rate reached a maximum of 28% at 55 DAS (Figure 2).
The other perennial species, A. tenuifolia, that under field conditions produces a very low number of viable seeds, which in pots poorly germinated mainly from the soil surface (data not shown). A. artemisiifolia achenes which are bigger than those of other tested Ambrosia species, were less affected by the depth of seeding and germinated even from 4 cm depths (Figure 3). A. artemisiifolia achenes emergence rate from the soil surface was higher in high temperatures regimes (28/34 • C, 22/28 • C) as compared to lower temperatures regimes (10/16 • C and 16/22 • C), and started 6 days later than in the higher temperature regimes (Figure 4).

Ambrosia Plant Development
Ambrosia species respond differently to different temperature regimes. Stem elongation differences were most evident in A. confertiflora as compared to A. tenuifolia and A. artemisiifolia, particularly at the two highest temperature regimes (Figure 5). Hence, at 100 DAS and 28/34 • C regime A. confertiflora plants reached an impressing height of more than 180 cm (average height of 140 cm), whereas at the lowest temperature regime (10/16 • C), the stem elongation was dramatically inhibited remaining at the rosette stage. Plant height of A. tenuifolia and A. artemisiifolia was less affected by the temperature regimes (Figure 5). A. tenuifolia plants were significantly shorter only at the lowest temperature regime, while in the A. artemisiifolia there was almost no effect to changes in temperature. A. confertiflora and A. artemisiifolia plants reach their maximal height at 60-80 DAS while A. tenuifolia maximal elongation was recorded at 100 DAS (data not shown).
The perennial species (A. confertiflora and A. tenuifolia), did not flower under long-day conditions throughout the current study. In contrary, flowering and fruit set were recorded in A. artemisiifolia as early as 25 DAS, in spite the fact that it is a shortday plant (Ziska et al., 2011), perhaps due to the reported short juvenile period in which it is not affected by day length (Deen et al., 1998).
Shoot biomass of the annual A. artemisiifolia was not affected by temperature, whereas the shoot biomass of A. confertiflora and A. tenuifolia was significantly reduced by the coldest temperature regime (Figure 6).
In the perennial species, underground biomass (roots and rhizomes) was not affected by the temperature regimes, while the underground biomass (roots only) of the annual species was dramatically increased when grown under the coldest regime (10/16 • C) (Figure 7). Thus, in order to elucidate the effect of temperature on resource allocation, we compared the effect of temperature on the ratio of underground/aboveground biomasses of the different species. The data presented in Figure 8 demonstrate that all tested species directed more resources to the underground organs as temperatures decline. This phenomenon was most evident in the annual A. artemisiifolia grown in the coldest conditions (10/16 • C) that diverted 6-fold more resources to the roots than to the shoot. For A. confertiflora and A. tenuifolia the temperature effect was more moderate: the ratio was 2-3 in the coldest temperature regime and decreased to 0.5 in the warmest temperature regime (Figure 8).

Sprout Development
Sprouts develop from rhizomes in the perennial species A. confertiflora (Figure 9) and A. tenuifolia only, and not in the annual A. artemisiifolia. In the net-house experiments conducted during the summer and fall (August-November),   sprouts first emerged from soil 25 DAS in both species. The average accumulated number of A. tenuifolia sprouts that developed from one mother plant at the end of the experiment (111 DAS) was five-fold higher than the number of sprouts that developed in A. confertiflora, 250 and 50, respectively (Figure 10).

DISCUSSION
This research elucidates the effect of temperature regimes on the early development and growth of three Ambrosia species that are in various distribution-status in Israel. The depth of seed placement in the soil strongly affects the germination and emergence of the three species: the seeds germinate well from the soil surface and up to 2 cm depth, as found also by Martínez et al. (2002). As these small seeds are placed deeper than 2 cm their emergence rate decreases accordingly (Farooq et al., 2019). The germination of A. confertiflora seeds are exceptionally sensitive to burial depth and they hardly emerge from more than 1 cm depth, while the bigger seeds of A. artemisiifolia possess more resources enabling them to emerge from deeper soil levels (2-4 cm), but as reported earlier they cannot emerge from depths >10 cm (Bassett and Crompton, 1975;Guillemin and Chauvel, 2011;Farooq et al., 2019). In contrast with other reports (Bazzaz, 1970(Bazzaz, , 1979Willemsen and Rice, 1972;Willemsen, 1975;Baskin, 1977, 1980;Leiblein-Wild et al., 2016), our germination studies showed that there is no need for stratification or other procedure to break the dormancy of A. artemisiifolia.
The two invasive species, A. confertiflora in Israel and A. artemisiifolia in Europe, demonstrate a higher phenotypic plasticity response to different temperatures, a common trait of invasive species. The effect of different temperatures on A. confertiflora and A. artemisiifolia seed germination is inverse: high temperatures reduce A. confertiflora germination rate and percentage, while the opposite occurs in A. artemisiifolia seeds as reported also by Gentili et al. (2019). As an annual plant, A. artemisiifolia is dependent mainly on the efficacy of seed germination and emergence for further favorable vegetative and reproductive growth (Gentili et al., 2019). The three examined Ambrosia species exhibit retarded growth in response to low temperatures, but A. confertiflora response to these low temperatures is the most extreme, leading to an increase in underground organs build-up, rosette formation and sprouting, while as temperature increases the stem elongates but the leaf number remains unchanged.
The two invasive A. artemisiifolia and A. confertiflora species have an interesting strategy of energy allocation following low temperatures to their underground plant parts (roots in A. artemisiifolia and rhizomes in A. confertiflora). In both A. confertiflora and A. artemisiifolia underground biomass increases 2 and 6-folds, respectively, in comparison to their above-ground mass. This ensures that plant resources are directed for supporting the establishment in soil under cold temperature periods. The annual short-day A. artemisiifolia plant grew rapidly, produced inflorescences and flowered 25 DAS, whereas A. confertiflora, did not flower throughout the four-month-long experiment. The high necessity to flower is not surprising as flowering is the sole way of reproduction of this annual plant, whereas the perennial Ambrosia species that reproduces also by sprouting is less dependent on seed production.
A. tenuifolia is not an invasive plant in Israel albeit the very high number of sprouts it rapidly produces. This is probably due to its less aggressive development in comparison to A. confertiflora and the fact that it produces a very low number of seeds, probably due to selfincompatibility (Friedman and Barrett, 2008). Furthermore, the seeds that are produced lack spines and are not dispersed by animals.
The naturalized A. artemisiifolia species is rapidly invading and spreading in Europe, Asia, and Australia. This annual invasive species successfully competes with local plant species due to its resistance to abiotic stresses, and its effective production and dispersion of viable achenes. Despite the sporadic detection of this species in Israel, A. artemisiifolia does not survive due to the dry summer in which the FIGURE 10 | Accumulated number of A. confertiflora and A. tenuifolia rhizome sprouts that developed over time in pots in the net house from August to November, 111 DAS. Error bars represent the Standard Error of the means. seedlings dry up and die. These plants can fit themselves to reduced moisture (Leiblein-Wild et al., 2014), but not to complete dryness such as the arid conditions in the Israeli summer. In highly moist conditions as in the reclaimed Hula Valley natural reserve in North-East of Israel, A. artemisiifolia grows rapidly and produces viable achenes, endangering the natural reserve and the neighboring irrigated agricultural fields (Yair et al., 2019).
The A. confertiflora species was first recorded in Israel in 1990 (Danin, 1994) and underwent a population explosion in less than 15 years (Yair et al., 2019), thus shifting from a naturalized to an invasive plant status. Several plant characteristics are responsible for this dramatic shift: fecundity, with the production of a vast number of viable achenes, easily caught on animal fur and other objects; ease of dispersion by man, farm equipment and the transportation of contaminated soil as well as rapid and long-distance seed transportation through water canals, streams and rivers (Yair et al., 2017). Additionally, rapid growth and fast sprouting contribute to its high competition with neighboring plants, resistance to temperature stress and growth plasticity may explain its rapid distribution and its establishment on river banks and fields adjacent to waterways.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

AUTHOR CONTRIBUTIONS
BR, YY, and HE conceptualized and designed the overall study. YY, YG, and MS conducted the experimental work. YY, YG, MS, HE, and BR analyzed the data, wrote, and edited the manuscript. All authors contributed to the article and approved the submitted version.

FUNDING
The research was partially funded by fund No. 45-01-003 of the Chief Scientist of the Ministry of Agriculture.