Introduction
Tropical ageratum (Ageratum conyzoides L.) is an annual herb belonging to the Asteraceae family and is considered a troublesome weed in tropical and subtropical regions across the world. The name Ageratum is derived from the Greek word ageras, which translates to “non-aging,” referring to the longevity of the entire plant (Kaur et al. Reference Kaur, Batish, Kohli and Singh2012). Ageratum conyzoides is commonly referred to as “billy goat weed” in Australia due to its peculiar smell, resembling that of a male goat (Okunade Reference Okunade2002). Ageratum conyzoides is known to have originated from Central America and the Caribbean (Ming Reference Ming1999). However, it has become widely established in warm regions across the globe, potentially due to its highly invasive nature. This weed poses a problem in agroecosystems, as it interferes with 36 crops in 46 different countries (Kaur et al. Reference Kaur, Kaur, Singh, Datta, Chauhan, Ullah, Kohli and Batish2023).
In Europe, the commercial cultivation of A. conyzoides began as an ornamental plant, and it was known to be cultivated in Bosnia, Bulgaria, France, and Sweden until the early 1970s (Johnson Reference Johnson1971). Although A. conyzoides was initially cultivated in Fiji and Lord Howe Island, Australia, it has now become a problematic weed species in the Australian farming systems (Kaur et al. Reference Kaur, Batish, Kohli and Singh2012). It is widely naturalized in humid parts of Northern Australia and has been identified as a major invasive weed in crops, pastures, and disturbed sites in northern Queensland and the Northern Territory (Holm et al. Reference Holm, Plucknett, Pancho and Herberger1977). While growing in association with desired vegetation, it can act as a host for various pathogens in numerous crops, including yellow leaf curl in tomato (Solanum lycopersicum L.), leaf curl in cotton (Gossypium hirsutum L.), and ringspot virus in papaya (Carica papaya L.) (Kaur et al. Reference Kaur, Kaur, Singh, Datta, Chauhan, Ullah, Kohli and Batish2023). Furthermore, this weed species is responsible for acting as a host for the chlorosis virus in capsicum (Capsicum annuum L.) (≤92% infection) in eastern Queensland, Australia (Sharman et al. Reference Sharman, Thomas, Tree and Persley2020).
Ageratum conyzoides can grow up to 1 m, displaying erect branching, trichome-covered stems, and a shallow taproot system (Kohli et al. Reference Kohli, Batish, Singh and Dogra2006). Due to the ability of A. conyzoides to form dense stands and produce large leaves in a wide range of temperatures, moisture conditions, soil textures, and altitudinal ranges, it can exploit available ecological niches, leading to rapid establishment by displacement of native vegetation (Kaur et al. Reference Kaur, Kaur, Singh, Datta, Chauhan, Ullah, Kohli and Batish2023). Ageratum conyzoides can produce a significant number of viable seeds, about 95,000 seeds plant−1; therefore, it is regarded as a prolific seed producer, with an extended seed-shattering period lasting from 5 to 8 mo (Ekeleme et al. Reference Ekeleme, Akobundu, Isichei and Chikoye2000; Rodriguez and Cepero Reference Rodriguez and Cepero1984). Nonetheless, these seeds remain viable for ≤12 mo (Okunade Reference Okunade2002). Seeds of A. conyzoides are minuscule (3.4 by 0.3 mm), positively photoblastic (Kaur et al. Reference Kaur, Batish, Kohli and Singh2012), and feature two prominent awns that facilitate anemochory (wind dispersal) and zoochory (animal dispersal). Additionally, the seeds can adhere to agricultural machinery and disperse within agroecosystems.
Seed germination and emergence are crucial events that determine a weed’s survivability, growth, reproduction, and fecundity in agroecosystems. These processes are known to be regulated by various environmental factors, such as temperature, light, soil moisture, salinity, and burial depth (Baskin and Baskin Reference Baskin and Baskin1977). Differential germination responses between herbicide-susceptible and herbicide-resistant weed populations have been extensively reported (Babineau et al. Reference Babineau, Mathiassen, Kristensen and Kudsk2017; Desai and Chauhan Reference Desai and Chauhan2021; Kumar et al. Reference Kumar, Jha, Lim and Stahlman2018; Vila-Aiub et al. Reference Vila-Aiub, Neve, Steadman and Powles2005). These observed differences could be attributed to the resource trade-off in plant metabolism, enabling survival under both biotic (e.g., pest or disease) and abiotic (e.g., herbicide applications) stress (Dhanda et al. Reference Dhanda, Kaur, Chaudhary, Jugulam, Hunjan, Sangha and Bhullar2022). Babineau et al. (Reference Babineau, Mathiassen, Kristensen and Kudsk2017) reported distinctive germination between iodosulfuron-resistant (acetolactate synthase [ALS] inhibitors [Group 2]) and iodosulfuron-susceptible populations of common windgrass [Apera spica-venti (L.) P. Beauv.].
A comprehensive understanding of seed germination biology is essential for predicting cohort establishment during the growing season, which aids in developing sustainable management protocols for problematic weed species, such as A. conyzoides. The primary objective of the current study is to evaluate the germination response of metsulfuron-resistant (MR) and metsulfuron-susceptible (MS) populations of A. conyzoides. This study addresses two research questions: (1) What are the alternative herbicide options to manage MR A. conyzoides? (2) Do MR populations of A. conyzoides differ in their environment-mediated germination responses (e.g., temperature, moisture, soil pH, salinity, and burial depth)? This study aimed to contribute to the limited knowledge available about the germination behavior of A. conyzoides. The results from this study will assist weed researchers and agronomists in developing successful management strategies for A. conyzoides.
Materials and Methods
Seed Collection
Seeds of two A. conyzoides populations were collected from a sugarcane field (18.664°S, 146.152°E) and a roadside area (18.590°S, 146.195°E) in Ingham, QLD, Australia, on March 23, 2021. Mature inflorescences were cut using a sickle and immediately placed in individual paper bags, sorted according to their respective populations. Seeds were collected from multiple plants (at least 50 plants) across the fields and were then bulked to ensure population representativeness. The collected seed bags were stored at room temperature (25 ± 2 C) at the Weed Science Laboratory (27.5551°S, 152.3343°E) of the University of Queensland, Gatton, until further use.
Herbicide Screening
A pot experiment was conducted in a randomized complete block design (RCBD) with four replications involving 14 herbicide treatments plus a control and two A. conyzoides (Sugarcane and Roadside) at Gatton Research Farm of the University of Queensland, from September 2021 to January 2022. Ten seeds of each population were placed on the surface of a pot filled with commercial potting mix (Centenary Landscaping, Mt Ommaney, QLD, Australia). The sieved potting mix was manually spread over seeds to achieve a consistent depth of 0.2 cm. Upon emergence, seedlings were thinned to 5 plants per pot. All pots were irrigated as needed.
Fourteen different herbicide treatments were applied at the 10- to 12-leaf stage (Table 1) using a stationary research track sprayer (manufactured by Woodlands Road Engineering, Gatton, QLD, Australia) calibrated to deliver a 108 L ha−1 spray volume through a flat-fan nozzle (TeeJet Spraying Systems Co., Wheaton, IL, USA). After spraying, all pots were kept away from any water supply for 24 h to account for the rainfast period of the herbicides.
Table 1. Herbicides, their trade names, rates, and sites of actions and adjuvants used on two Ageratum conyzoides populations (Sugarcane and Roadside) collected from Ingham, QLD, Australia
a ALS, acetolactate synthase; EPSPS, 5-enolpyruvylshikimate-3-phosphate synthase; PPO, protoporphyrinogen oxidase.
b Uptake: Corteva Agriscience Australia Pty Ltd, Chatswood, New South Wales, Australia.
c Hasten: BASF Australia Ltd, Southbank, Victoria, Australia.
d Cando: Nufarm, Australia Ltd, Laverton North, Victoria, Australia.
The surviving seedlings per pot were recorded at 28 d after spraying (DAS) and converted into a survival percentage. At 28 DAS, surviving seedlings per pot were cut from the base, placed in paper bags, and then oven-dried at 70 C for 72 h. The oven-dried samples were weighed using a digital scale to record dry matter (g).
Metsulfuron Dose Response
The herbicide screening experiment revealed differential responses of the two populations to metsulfuron. Therefore, this pot experiment was conducted using RCBD with six replications and two A. conyzoides populations (Sugarcane and Roadside) from January to October 2022. Seed placement was carried out as described earlier. After emergence, plants were thinned to 4 plants per pot. Seven different doses of metsulfuron (0, 0.8, 1.5, 3, 6, 12, and 24 g ai ha−1) were applied using a stationary research track sprayer, calibrated to deliver a 108 L ha−1 spray volume through flat-fan nozzles (TeeJet® XR 110015). Importantly, 3 g ai ha−1 is the recommended dose. Plants were sprayed at the 4- to 6-cm or 4- to 6-leaf stage. Sprayed plants were kept away from any water supply for 24 h to ensure the required rainfast period. Survival percentage and dry matter were recorded as described earlier.
Germination Ecology Experiments
Temperature-mediated Germination Response
An experiment was conducted in the Weed Science Laboratory, Gatton, QLD, Australia, by using RCBD with three replications from September to November 2022. Twenty-five seeds were evenly distributed in a petri dish (92-mm diameter by 16-mm height) with double filter papers (Macherey-Nagel, Duren, Germany). The petri dishes were moistened with 5 ml of ionized water using a micropipette (Boeco, Hamburg, Germany). Petri dishes were then placed in sealable transparent plastic bags, which were then placed in five different incubators (Labec Laboratory, Marrickville, Sydney, New South Wales, Australia) set to deliver distinct thermal conditions, 15/5, 20/10, 25/15, 30/20, and 35/25 C with a 12/12-h day/night photoperiod for 28 d. This experiment was repeated 10 d after completion of the first run.
Salinity-mediated (NaCl) Germination Response
An experiment was carried out using RCBD with three replications and two A. conyzoides populations (Sugarcane and Roadside), from October to December 2022. Seed placement in petri dishes was carried out as described previously. Petri dishes were treated with 0 (water only), 20-, 40-, 80-, 160-, and 320-mM sodium chloride (NaCl) solutions using a micropipette corresponding to the treatments. Petri dishes, including control (0 mM NaCl), were placed in the incubator set at 30/20 C (temperature conditions found optimum in the temperature experiment) and 12/12-h photoperiod for 28 d. This experiment was repeated 10 d after completion of the first run.
Osmotic Potential–mediated Germination Response
An experiment was undertaken using RCBD with three replications and two A. conyzoides populations (Sugarcane and Roadside) from October to December 2022. Seed placement in petri dishes was done as described earlier. All petri dishes were either treated with ionized water (control: 0 MPa osmotic potential) or −0.1, −0.2, −0.4, −0.8, and −1.6 MPa osmotic potential solutions corresponding to the treatment. The osmotic potential solutions were prepared by dissolving the appropriate quantity of polyethylene glycol (PEG) powder (Equation 1) in deionized water (Michel and Radcliffe Reference Michel and Radcliffe1995):
$${\rm{\Psi s}} = 0.13{({\rm{PEG}})^{2\;}}T - 13.7{({\rm{PEG}})^{2\;}}$$
where ψs represents gravimetric component, “PEG” represents the quantity of polyethylene glycol (g PEG g−1 water) and T represents the temperature. The PEG-treated petri dishes plus control were incubated at 30/20 C temperature and 12/12 h-photoperiod for 28 d. This experiment was repeated 10 d after completion of the first run.
Response Variable Recorded in Germination Ecology Experiments
Once a clearly visible radicle emerged (≥1 mm), the seeds were considered germinated, and germination counts were recorded by removing seeds with forceps at 28 d after incubation. The germination counts were then converted into germination (%), which was used for further analysis.
Burial Depth–mediated Emergence Response
A pot experiment was conducted in RCBD with three replications and two A. conyzoides populations (Sugarcane and Roadside) from October 2022 to January 2023 at Gatton Research Farm of the University of Queensland. Thirty seeds of A. conyzoides were sown in individual pots (14-cm diameter) at a depth of 0 (on the soil surface), 0.5, 1, 2, and 4 cm for the emergence study. The soil used in this experiment was collected from the Gatton Research Farm and sieved through a 0.3-cm mesh. The soil is identified to contain 33:46:21 of sand:silt:clay with 2.6% organic matter and 7.1 pH. The pots were subirrigated to avoid damage caused by the sprinkler system. The seedlings were considered emerged once their coleoptiles became visible on the soil surface. The emergence counts were recorded after 28 d and converted into the emergence percentage before being subjected to data analysis. This experiment was repeated 21 d after completion of the first run.
Statistical Analyses
RStudio (v. 4.2.2, Posit Software, Boston, MA, USA) was the primary tool used for statistical model development. Two-way ANOVA models were developed for each experiment using the aov function to assess the variation between two runs. No significant difference was observed; therefore, data were pooled across the runs. Data from herbicide screening and temperature-mediated germination experiments were square-root transformed before being subjected to further analysis, as normality violations were detected in diagnostic plots developed using the DHARMa package. These data were fit to the linear mixed model using the lme4 package, and the effects of explanatory variables were compared based on the type II Wald χ2 test using the Anova function of the car package. For the herbicide screening experiment, herbicide treatments, populations, and their interaction were fixed effects, and replication was a random effect. For the temperature-mediated germination experiment, temperatures, populations, and their interaction were fixed effects, and replication was a random effect. Means were separated by Fisher’s protected LSD test (P < 0.05) using the emmeans and multcomp packages.
Metsulfuron dose–response, NaCl, osmotic potential, and burial depth experimental data were fit to a three-parameter log-logistic model (Equation 2) using the drm function of the drc package in R Studio (Ritz et al. Reference Ritz, Baty, Streibig and Gerhard2015; Seefeldt et al. Reference Seefeldt, Jensen and Fuerst1995):
$$Y = {\rm{\;}}\left\{ {d{/}1 + \exp \left[ {b{\rm{\;}}\left( {{\rm{log}}X - {\rm{\;}}{{\rm{E}}_{50}}} \right)} \right]} \right\}$$
where Y is the response variables (survival [%], dry matter [% of control], germination [%], or emergence [%]); d represents the upper limit; b is the slope of each curve; X is the metsulfuron dose (g ai ha−1), NaCl concentrations (mM), osmotic potential (MPa), or burial depth (cm); and E50 represents the metsulfuron dose (g ai ha−1) required to cause 50% injury (referred to as LD50 and GR50 for survival percentage and dry matter, respectively), NaCl concentration (mM) or osmotic potential (MPa) required to inhibit 50% germination (referred to as GI50), or burial depth (cm) required to inhibit 50% emergence (referred to as EI50).
In addition, the metsulfuron dose required to cause 10% and 90% injury (referred to as LD10/GR10 and LD90/GR90, respectively), NaCl concentration (mM) and osmotic potential (MPa) required to inhibit 10% and 90% of germination (referred to as GI10 and GI90, respectively), and burial depth (cm) required to inhibit 10% and 90% emergence (referred to as EI10 and EI90, respectively) were calculated using the ED function of the drc package. For the metsulfuron dose–response experiment, estimated parameters were compared by approximate t-test using the CompParm function of the drc package. The resistant index (R:S) was calculated by dividing LD50 and GR50 of two A. conyzoides populations.
Results and Discussion
Herbicide Screening
The results showed that the interaction of herbicides and populations had a significant impact (P = 0.001) on both dry matter (g) and survival percentage (Table 2). Paraquat, saflufenacil, saflufenacil + trifludimoxazin, glufosinate, and tiafenacil were found to be effective in controlling both populations of A. conyzoides (100% dry matter reduction and 0% survival at 28 DAS) (Table 2). The application of 2,4-D, 2,4-D + picloram, and glyphosate provided >95% dry matter reduction of both populations at 28 DAS; however, the survival percentage ranged from 10% to 38% (Table 2).
Table 2. Survival and dry matter of two populations of Ageratum conyzoides (Sugarcane and Roadside) 28 d after herbicide treatments at the University of Queensland, Gatton, Australia
a Means followed by similar letters within the same response variable are nonsignificant differences based on Fisher’s protected LSD test (P < 0.05). The interaction of herbicide and population were significant (P < 0.001) for both survival (%) and dry matter; therefore, Fisher’s protected LSD test was conducted for interactive (herbicide:population) variable.
b Values in parentheses are dry matter reduction percentages compared with the control.
Metsulfuron provided varied degrees of control on two populations, as the Sugarcane population displayed 98% survival as opposed to 0% survival of the Roadside population (Table 2). These results suggest that the Sugarcane population could be resistant to metsulfuron; therefore, the dose–response study was conducted on both populations to confirm the level of resistance.
Metsulfuron Dose Response
The survival and dry matter responses of Sugarcane and Roadside populations were accurately described (P > 0.05) by a three-parameter log-logistic model (Equation 2) (Figure 1A and 1B). The Sugarcane population exhibited a 100% survival rate until the metsulfuron dose increased to 12 g ai ha−1. To achieve complete control of the Sugarcane population, 48 g ai ha−1 metsulfuron was required, 16 times more than the recommended rate. In contrast, a mere 1.5 g ai ha−1 of metsulfuron was sufficient to control the Roadside population (Figure 1A and 1B). The LD50 of the Sugarcane population (26.9 g ai ha−1) was significantly higher (P < 0.001 based on an approximate t-test) compared with the LD50 of the Roadside population (0.5 g ai ha−1) (Table 3). Based on LD50 values of both populations, the Sugarcane population exhibited an average 54-fold (R/S ratio) higher resistance to metsulfuron compared with the Roadside population (Table 3). Similarly, the GR50 of the Sugarcane population (3 g ai ha−1) was significantly higher (P < 0.001 based on an approximate t-test) than that of the Roadside population (0.1 g ai ha−1), indicating that the Sugarcane population was 30-fold (R/S ratio) more resistant to metsulfuron compared with the Roadside population (Table 3). Additionally, a consistent trend was observed in LD10/GR10 and LD90/GR90 values (Table 3). The metsulfuron resistance in the Sugarcane population may be attributed to the extensive use of metsulfuron in agroecosystems, as the Roadside population was found to be susceptible. To our knowledge, this is the first report of MR A. conyzoides in Australia, as no other cases have been reported (Heap Reference Heap2023).
Figure 1. Dry matter (A) and survival frequency (B) of metsulfuron-susceptible (Roadside) and metsulfuron-resistant (Sugarcane) populations of Ageratum conyzoides at 28 d after the application of increasing doses of metsulfuron, modeled by a three-parameter log-logistic regression (Y = {d/1 + exp[b(logX − E50)]}) model. The parameters of the fitted model are presented in Table 3.
Table 3. Three-parameter log-logistic regression (Y ={d/1 + exp[b(logX − E50)]}) parameters for survival frequency and dry matter of metsulfuron-susceptible (Roadside) and metsulfuron-resistant (Sugarcane) populations of Ageratum conyzoides mentioned in Figure 1
a Regression parameters: b, slope; LD10 and GR10, lethal dose of metsulfuron (g ai ha−1) required for 10% control and dry matter reduction compared with control, respectively; LD50 and GR50, lethal dose (g ai ha−1) of metsulfuron required for 50% control and dry matter reduction compared with control, respectively; LD90 and GR90, lethal dose (g ai ha−1) of metsulfuron required for 90% control and dry matter reduction compared with control, respectively.
b Resistance/susceptible ratio is calculated based on LD50 and GR50 of Sugarcane and Roadside populations.
The evolutionary process of resistance development results in functional changes in weeds (Baucom and Mauricio Reference Baucom and Mauricio2004) that aid in their survival and reproduction under abiotic stress (e.g., herbicide application). However, these spontaneous functional changes could reduce the overall fitness of weeds (Roux et al. Reference Roux, Gasquez and Reboud2004; Tardif et al. Reference Tardif, Rajcan and Costea2006), termed “fitness penalties.” As an illustrative example from another system, early germination has been observed in sulfonylurea-resistant kochia [Bassia scoparia (L.) A.J. Scott] when compared with a sulfonylurea-susceptible population (Thompson et al. Reference Thompson, Thill and Shafii1994). Therefore, in the present study, Sugarcane (MR) and Roadside (MS) populations were further employed to evaluate the germination and emergence responses in different environmental conditions.
Germination Ecology Experiments
Temperature-mediated Germination Response
The temperature regimes significantly impacted (P < 0.001) the germination of A. conyzoides; however, no significant difference was observed (P = 0.844) in germination between Sugarcane (MR) and Roadside (MS) populations (Figure 2). Therefore, the population variable was removed from the fitted model to select the parsimonious model. Seeds of both populations germinated at a wide range of temperatures (from 15/5 to 35/25 C) (Figure 2), indicating the potential for A. conyzoides to germinate throughout the year in the Queensland area. Similar year-round germination abilities were found in redflower ragleaf [Crassocephalum crepidioides (Benth.) S. Moore], horseweed [Conyza canadensis (L.) Cronquist], and A. conyzoides (Yuan and Wen Reference Yuan and Wen2018). The highest germination (>90%) was recorded at 30/20 C, which could be considered an ideal temperature for optimum germination. Moreover, A. conyzoides exhibited >80% germination at 20/10, 25/15, and 35/25 C (Figure 2). These results suggest that A. conyzoides could germinate throughout the spring and summer months and may possibly emerge in the early autumn in the northern part of Australia. The interference of this weed species in summer months with vegetable crops, such as S. lycopersicum, potato (Solanum tuberosum L.), C. annuum, and eggplant (Solanum melongena L.), has been observed (Pegg and Moffett Reference Pegg and Moffett1971). Ageratum conyzoides also germinated (25% to 30%) at low temperatures (15/5 C); however, winter frost may damage seedlings in the early vegetative phase. We also acknowledge that the photoperiod that was utilized in the experiment does not reflect the photoperiod when temperatures decrease. Therefore, a phenology study on A. conyzoides should be conducted in the future, wherein different populations can be planted every month throughout the year to explore the survivability, growth, and fecundity of this weed species.
Figure 2. Germination response of Ageratum conyzoides when incubated in five alternating temperature regimes with 12/12-h light/dark photoperiod. The interaction of alternating temperature regimes and populations was nonsignificant; hence, two populations were combined to perform the mean separation test. Similar letters denote nonsignificant difference based on Fisher’s protected LSD test (α = 0.05). The boxes, center lines, and vertical lines represent the standard error of mean, mean, and whole data range, respectively.
Salinity-mediated (NaCl) Germination Response
The germination of A. conyzoides was significantly affected (P <0.001) by salinity but not by population (P = 0.233). Therefore, the population variable was removed from the fitted model to select the parsimonious model. The three-parameter log-logistic model adequately described (P > 0.05) the germination of A. conyzoides in response to salinity levels (Figure 3A). The maximum germination (88% to 100%) was observed in 0 mM NaCl concentration (control), which declined gradually as the salinity level increased (Figure 3A). Germination was completely inhibited at the 320 mM NaCl concentration (Figure 3A). The GI10, GI50, and GI90 were 96.45, 168.84, and 295.8 mM, respectively (Table 4).
Figure 3. Effect of increasing levels of sodium chloride (NaCl) concentration (A), osmotic potential (B), and seed burial depth (C) on germination or emergence of Ageratum conyzoides, modeled by a three-parameter log-logistic regression model (Y = {d/1 + exp[b(logX − E50)]}). The parameters of the fitted model are presented in Table 4.
Table 4. Three-parameter log-logistic regression (Y ={d/1 + exp[b(logX − E50)]}) parameters for germination in sodium chloride (NaCl) concentrations and osmotic potential and emergence in seed burial depth of Ageratum conyzoides mentioned in Figure 3
a b, slope; GI10, NaCl concentration and osmotic potential required for 10% germination inhibition; GI50, NaCl concentration and osmotic potential required for 50% germination inhibition; GI90, NaCl concentration and osmotic potential required for 90% germination inhibition. Similarly, EI10, burial depth required for 10% emergence inhibition; EI50, burial depth required for 50% emergence inhibition; EI90, burial depth required for 90% emergence inhibition.
b R2, coefficient of determination averaged over two populations within the same ecological factors.
Differential germination responses to salinity levels have been reported in two Asteraceae weeds: eclipta [Eclipta prostrata (L.) L.] and siamweed [Chromolaena odorata (L.) R.M. King & H. Rob.]. The GI50 values of E. prostrata and C. odorata were 194 mM (Chauhan and Johnson Reference Chauhan and Johnson2008b) and 171.6 mM (Chauhan and Johnson Reference Chauhan and Johnson2008a), respectively. Due to the high salinity tolerance of these Asteraceae weeds, crop cultivation might be impacted not only by salinity but also by weed competition (Chauhan and Johnson Reference Chauhan and Johnson2008a). A similar study on beggar’s tick (Bidens pilosa L.) in Queensland reported that 120 mM NaCl inhibited 50% germination, while no seeds germinated at 250 mM NaCl (Chauhan et al. Reference Chauhan, Ali and Florentine2019). These results indicated that A. conyzoides might be more immune to salinity compared with other Asteraceae weed species. According to the Australian Bureau of Statistics, 107,000 ha of land in Queensland is considered saline (>100 mM NaCl) (Queensland Government 2013). Due to the higher salinity tolerance, these saline conditions could provide a germination niche for the establishment of this weed species.
Osmotic Potential–mediated Germination Response
The osmotic potential significantly affected (P < 0.001) the germination of A. conyzoides; however, no significant difference (P = 0.716) was found in germination between MR and MS populations. Therefore, the population variable was excluded from the fitted to select the parsimonious model. Germination of A. conyzoides under various levels of water stress was modeled using a three-parameter log-logistic model (P > 0.05) (Figure 3B). Maximum germination (92% to 100 %) was recorded under 0 MPa osmotic potential (Figure 3B). Consistent with the salinity-mediated germination response, germination declined as osmotic potential decreased (Figure 3B). The osmotic potential of −0.8 MPa was required to completely inhibit the germination of A. conyzoides (Figure 3B; Table 4).
At −0.3 MPa osmotic potential, the germination of A. conyzoides was 50% inhibited (GI50) (Table 4), which is lower than the GI50 of blue weed (Ageratum houstonianum Mill.) (−0.7 MPa) (Lamsal et al. Reference Lamsal, Devkota, Shrestha, Joshi and Shrestha2019). In other Asteraceae weed species, such as C. odorata, the osmotic potential required to inhibit 50% germination was recorded at −0.6 MPa, which is also higher than the GI50 of A. conyzoides. These results suggest that A. conyzoides is less tolerant to water stress compared with other Asteraceae weed species. Moreover, in the present study, the germination of A. conyzoides drastically declined beyond the −0.4 MPa osmotic potential, suggesting that a moist environment is needed for the germination of this weed species. Lal (Reference Lal2022) mentioned that the occurrence of A. conyzoides is more common in tropical and subtropical regions with higher precipitation, which supports the aforementioned claim of the less water stress–tolerant characteristic of this weed species.
Burial depth–mediated Emergence Response
The burial depth had a significant effect (P < 0.001) on the germination of A. conyzoides. However, there was no significant difference (P = 0.840) in germination between the MR and MS populations. Therefore, the population variable was removed from the fitted model to select the parsimonious model. The three-parameter log-logistic model accurately described (P > 0.05) the emergence response of A. conyzoides under increasing burial depths (Figure 3C). The maximum emergence (66.7% to 93.3%) was observed for surface seeds (Figure 3C). However, with increasing burial depth, the emergence of A. conyzoides drastically declined and was completely inhibited at 1-cm depth (Figure 3C). The burial depth values required to inhibit 10% (EI10), 50% (EI50), and 90% (EI90) emergence were 0.3, 0.4, and 0.5 cm, respectively (Table 4).
As A. conyzoides populations failed to emerge from 1-cm or greater burial depth (Figure 3C), this species is categorized as a surface-emerging weed. A similar emergence pattern has been reported for other Asteraceae weed species, such as A. houstonianum (Lamsal et al. Reference Lamsal, Devkota, Shrestha, Joshi and Shrestha2019) and C. canadensis (Nandula et al. Reference Nandula, Eubank, Poston, Koger and Reddy2006). These results suggest that the seeds of A. conyzoides are photoblastic in nature. Due the absence of sunlight in a deeper soil profile, seeds were not able to emerge from the 1-cm burial depth. In conventional tillage systems, A. conyzoides seeds can be buried below 1 cm through tillage to manage this weed species. Ageratum conyzoides seeds can remain viable for up to 12 mo in the soil seedbank (Marks and Nwachuku Reference Marks and Nwachuku1986). Therefore, tillage operations should be carried out at a shallow depth to avoid the reintroduction of seeds from a deeper soil profile (>1 cm) to the soil surface.
The present study discovered the evolution of metsulfuron resistance in an A. conyzoides population (MR) collected from a sugarcane field in Ingham, QLD, Australia. However, the germination and emergence studies revealed a similar response between the MR and MS populations to various environmental factors. The germination ability of A. conyzoides under a wide range of alternating temperature regimes indicates the potential year-round occurrence of this weed species in the Queensland region. Ageratum conyzoides is well adapted to a broad range of salinity levels; however, its water-stress adaptability is lower compared with other Asteraceae weed species. Ageratum conyzoides could not emerge when buried at 1 cm or deeper in the soil, suggesting the potential utilization of shallow tillage to manage this weed species. This study identifies important research gaps, such as a lack of phenology studies, to acquire comprehensive knowledge of the year-round growth, reproduction, and fecundity of this weed species. The evolution of metsulfuron resistance impacted neither the salinity and water-stress tolerance nor the emergence adaptability to different burial depths of A. conyzoides. However, we acknowledge that these results cannot be generalized, as we used only one population of each category and our findings may not apply to other weed species. Results from the current study could be useful for weed scientists and agronomists to proactively design an effective management protocol to control this weed species.
Funding
This research received no specific grant from any funding agency or the commercial and not-for-profit sectors.
Competing interests
The first and second authors are married to each other. No professional competing interests exist.
Open access
