Evaluation of Organic Options for Johnsongrass (Sorghum halepense) Control during Winter Fallow

Abstract Johnsongrass [Sorghum halepense (L.) Pers.] is one of the most problematic perennial grass weed species in row-crop production across the southern United States. Control of this species is especially challenging in organic systems due to a lack of effective options. A field experiment was conducted at the Texas A&M research farm near College Station, TX, from fall 2019 to spring 2021 to evaluate various nonchemical options for managing S. halepense in the fallow season, implemented over 2 yr in the same locations. The treatments included disking once, disking twice, disking + immediate flooding, disking + flush irrigation + flooding, disking twice + flooding after the first frost, periodic mowing, acetic acid treatment, and disking + tarping. Disking + immediate flooding, disking + flush irrigation + flooding, and disking + tarping were the most effective treatments. Compared with the nontreated control plots, these treatments reduced S. halepense aboveground density (<9 plants m–2 vs. 64 plants m–2), aboveground biomass (<80 g m–2 vs. 935 g m–2), rhizome biomass (<4 g m–2 vs. 55 g m–2), rhizome node number (<25 nodes m–2 vs. 316 nodes m–2), and rhizome length (<42 cm m–2 vs. 660 cm m–2). Disking twice + flooding after the first frost did not show a consistent impact. Periodic mowing also reduced S. halepense density (12 plants m–2 vs. 64 plants m–2) and other variables compared with the control plots at the end of the study in spring 2021. Disking alone once or twice each growing season or repeated application of acetic acid failed to control S. halepense. These results indicate that well-timed nonchemical management practices such as tarping and flooding implemented during the winter fallow can be very effective in reducing S. halepense densities.


Introduction
Johnsongrass [Sorghum halepense (L.) Pers.], a perennial summer grass weed native to the Mediterranean region, is one of the most problematic weed species across the southern United States (McWhorter 1993;McWhorter and Hartwig 1972;Ohadi et al. 2018;Travlos et al. 2018).Several biological characteristics of S. halepense contribute to its invasiveness (McWhorter 1961;Taylorson and McWhorter 1969;Warwick et al. 1986).This species has a remarkable ability to reproduce both sexually (seeds) and asexually (rhizomes) (Anderson et al. 1960;Horowitz 1972aHorowitz , 1973)).A single S. halepense plant can produce up to 80,000 seeds (Anderson 1996;Monaghan 1979) that can remain viable in the soil for up to 6 yr (Horowitz 1973).
Seedling S. halepense (i.e., from seeds) can begin forming rhizomes approximately 3 wk after emergence and can produce about 65 m of rhizomes in 5 mo of growth (McWhorter 1961).The rhizomes store carbohydrate reserves and serve as overwintering structures; the axillary and terminal buds produce new vegetative shoots and contribute to reestablishment in the spring (Anderson et al. 1960;McWhorter 1961;Monaghan 1979;Travlos et al. 2019).Some S. halepense buds may remain inactive/dormant during environmental stress conditions and regrow when conditions improve (McWhorter 1961).Rhizomes sprout in soil ranging from 15 to 30 C (Hull 1970), whereas seed germination occurs at 20 to 35 C (Taylorson and McWhorter 1969).In southeast Texas, S. halepense rhizomes can sprout as early as February (GCS and MB, personal observations).Although S. halepense rhizomes cannot survive when the soil temperatures reach below −9 C (McWhorter 1972a;Stoller 1977), winter temperatures in much of the southern United States are not low enough to kill the rhizomes.
Repeated mowing can reduce S. halepense rhizome growth and prevent seed production (Warwick and Black 1983).Other mechanical control techniques, such as disking and tillage, can effectively control young seedlings; however, fragmentation of rhizomes by tillage can further spread S. halepense, especially if sufficient soil moisture is available following tillage (Travlos et al. 2019).McWhorter and Hartwig (1965) reported that repeated disking or tillage operations could deplete rhizomes and reduce S. halepense stands as long as seedlings and sprouts are uprooted and rhizome segments are desiccated.The authors noted that multiple disking operations over the growing season effectively controlled S. halepense in soybean fields.
Flooding can be an effective integrated weed management tool (Price et al. 2010;Singh et al. 2006).Flooding impacts plants by creating an anaerobic environment detrimental to plant growth (Saini 2014).McWhorter (1971) reported that flooding effectively controlled S. halepense rhizomes, and the best results were obtained when the field was covered with 7 to 10 cm of water before rhizome sprouting.In another study, flooding soil with 5 to 10 cm of water for 1 to 2 wk effectively controlled all freshly planted S. halepense rhizomes in both greenhouse and field conditions (McWhorter 1972b).The use of tarping to suppress broadleaf and grass weeds has been well documented (Law et al. 2008;Ricotta and Masiunas 1991;Zhang et al. 2021), and solarization with clear tarping has been shown to effectively kill S. halepense rhizomes (Elmore et al. 1993;Stapleton 2012).However, no studies have evaluated black tarping for rhizomatous S. halepense control.
Evaluation of different nonchemical management tactics side by side allows for determining relative effectiveness under comparable conditions and helps make informed management decisions.The long fallow season following the harvest of the main cash crop in the southern United States can be utilized for aggressive management of rhizomatous S. halepense using nonchemical methods.However, there have been no studies evaluating S. halepense control in the fallow season.The objective of this research was to evaluate rhizomatous S. halepense control using a number of nonchemical methods implemented during the winter fallow season in southeast Texas.

Experimental Site
Field experiments were conducted at the Texas A&M University Research Farm near College Station, TX (30.552226°N, 96.424928°W ).The site is characterized by a Ships clay soil type (very fine, mixed, active, thermic Chromic Hapluderts).The experimental site had been managed organically for several years and was left fallow (the field was periodically mowed, but no cash crop was planted) for the 2 yr immediately before the experiment.There was a natural infestation of S. halepense at the experimental site in high and uniform densities (19 plants m −2 ) before the initiation of the experiment (Figure 1).

Treatment Details
The study was conducted from fall 2019 to late spring 2021, comprising two annual cycles, each running from late September to late April, where treatments were applied to the same plots across both years.Nine treatments were implemented in a randomized complete block design with three replications.The study was not repeated across time or location.Each plot measured 15.2-m long and 3.3-m wide.Flooded plots had a soil berm (50-cm tall) built around them.Flooding was accomplished by pumping well water through poly irrigation pipes.The treatments included: T1: An untreated control, which was mown once at the beginning of the experiment (late September) and left untouched throughout the experiment.T2: Disking once to a depth of 15 cm in late September.T3: Disking twice to a depth of 15 cm, once in late September and again after the S. halepense had regrown to 30-to 38-cm height in late October.T4: Disking once to a depth of 15 cm in late September, followed immediately by flooding for 14 d at a 20-cm depth (Figure 2A).T5: Disking once to a depth of 15 cm in late September, followed by flush irrigation to soil field capacity to encourage emergence/ sprouting, then flooding for 14 d at a 20-cm depth when S. halepense height was 2.5 to 5 cm.Flush irrigation was accomplished with the same poly pipes used for irrigation.T6: Disking twice to a depth of 15 cm, once in late September and again after the S. halepense had regrown to 30 to 38 cm height in late October, followed by flooding for 14 d at a 20-cm depth after the first frost by late November (plants were 20-to 30-cm tall, going into the dormant stage).T7: Periodic mowing when the plant growth reaches 30-to 38-cm tall.T8: Spraying acetic acid (Green Gobbler Concentrated Vinegar, Green Gobbler, Gurnee, IL) undiluted (30% concentration at 188 L ha −1 ) when S. halepense reached 30-to 38-cm tall (Abouziena et al. 2009).T9: Disking to a depth of 15 cm in late September, followed immediately by installation of black tarp (6-mil polyethylene) (Figure 2B).
The treatments were implemented in the same plots during the two study years to evaluate the impact of repeated applications of these treatments.A sunn hemp (Crotalaria juncea L.; Hancock Seed, Dade City, FL) summer cover crop was planted at 33.5 kg ha −1 seeding rate during the cropping season (May to September 2020) in the entire experimental area.The purpose of the cover crop was to simulate an intensive and competitive cash crop during the summer season.The sunn hemp cover was terminated on September 21, 2020, using a roller-crimper.The field was then disked before implementing the second cycle of treatments.The specific dates of field operations for the two study years are provided in Table 1.

Data Collection
Initial S. halepense density was determined before implementing the treatments each year.In 2019, when S. halepense was uniformly distributed in the experimental site, the average density for the entire field was calculated by randomly placing sixteen 1-m 2 quadrats throughout the field and counting the number of rhizomatous S. halepense shoots in each.Only the rhizomatous S. halepense shoots were monitored in this study, because the seedling S. halepense that established after fall is less likely to survive the winters in the study location; moreover, the rhizomatous shoots give a good representation of the effectiveness of the treatments.In 2020, densities before treatment implementation were recorded in each plot by placing two 50 cm by 50 cm quadrats and counting the number of rhizomatous shoots in each quadrat.In T5, S. halepense sprouting after flush irrigation was quantified in five 50 cm by 50 cm quadrats per plot before flooding.The final S. halepense densities for each cycle were recorded on April 17, 2020, and April 28, 2021, in all plots.In 2020, the number  of rhizomatous S. halepense shoots in each plot was recorded in three 50 cm by 50 cm quadrats, whereas in 2021, data were collected from five 50 cm by 50 cm quadrats.Additionally, at the termination of the experiment in late April 2021, S. halepense dry biomass and rhizome density were determined for each plot to assess the cumulative impact of the treatments after 2 yr of implementation.In two 50 cm by 50 cm quadrats per plot, all aboveground S. halepense biomass was harvested, dried for 7 d at 60 C, and weighed.In each of those quadrats, all rhizome segments were dug up for a depth of 15 cm.The total linear rhizome length was measured, the number of nodes was counted, and the samples were dried for 7 d at 60 C and weighed for rhizome biomass.

Statistical Analysis
Sorghum halepense aboveground densities were subjected to one-way repeated-measures ANOVA with the generalized linear mixed model (PROC GLIMMIX) in SAS (v.9.4, SAS Institute, Cary, NC), with data collection timings (spring of 2020, fall of 2020, and spring of 2021) as the repeated measures and each plot as the experimental unit.A heterogeneous autoregressive 1 [ARH(1)] covariance structure was chosen, as it provided the lowest corrected Akaike information criterion value.Treatment means were separated using Tukey's honestly significant difference (HSD) method at α = 0.05.The aboveground densities of S. halepense between fall 2019 and spring 2021 were compared for each treatment using the Student's t-test at α = 0.05 in SAS.
Sorghum halepense aboveground biomass, rhizome length, rhizome biomass, and rhizome node number were subjected to one-way ANOVA with the generalized linear mixed model (PROC GLIMMIX) in SAS.Treatment was considered to be the fixed effect, while replication was regarded to be the random effect.A link function (link = log) was used in the GLIMMIX model statement to address potential normality issues.Treatment means were separated using Tukey's HSD method at α = 0.05.

Results and Discussion
Our results demonstrated, in a side-by-side comparison, the efficacy of various nonchemical methods for controlling rhizomatous S. halepense.Varying levels of S. halepense control were observed among the treatments (Figure 3).Disking once (T2), disking twice (T3), disking once þ flooding (T4), and periodic mowing (T7) reduced S. halepense densities at the termination of the study in spring 2021 (61%, 54%, 89%, and 85%, respectively), but the treatment impacts were less prominent by spring 2020, just after the first cycle of treatments (Figure 3).Disking þ flooding for 14 d after the first frost (T6) reduced S. halepense densities in spring 2020, but the effect was inconsistent in the fall 2020 and spring 2021 observations.Disking þ flush irrigation þ flooding after S. halepense sprouting (T5) and disking þ installation of black tarp (T9) were the most effective treatments, consistently reducing S. halepense densities in all three observation timings (spring 2020, fall 2020, spring 2021).At the end of the study, T5 and T9 reduced S. halepense densities by 95% and 97%, respectively.Throughout the study, disking twice þ flooding at first frost (T6) and acetic acid application (T8) had the least impact on S. halepense densities (Figure 3).
For each treatment, the results of the t-tests compared the densities between fall 2019 and spring 2021 showed that disking at the initiation of the experiment þ flooding for 14 d (T4), disking at the initiation of the experiment þ flush irrigation þ flooding for 14 d (T5), periodic mowing (T7), and disking þ black tarping (T9) resulted in significantly lower S. halepense densities in spring 2021 compared with fall 2019, whereas all other treatments (T1, T2, T3, T6, and T8) had statistically comparable or higher S. halepense densities.
Results showed that disking alone once or twice was ineffective for controlling S. halepense.Disking cuts S. halepense rhizomes into smaller pieces, which encourages new shoot growth from each piece and leads to further spread (Horowitz 1972a;McWhorter and Hartwig 1965;Travlos et al. 2019).To achieve effective S. halepense control with tillage, frequent disking may be required  Weed Science (Johnson et al. 1997).In previous research, McWhorter and Hartwig (1965) reported that repeated disking 10 times at 4-to 6-d intervals effectively controlled S. halepense.Frequent disking and exposing the rhizomes to heat or cold can destroy them rapidly (McWhorter 1971), as rhizomes do not survive when exposed to temperatures below −4 C (Warwick et al. 1986) or above 30 C for more than 7 d under dry conditions (Warwick and Black 1983).The effectiveness of disking is also associated with the extent of rhizome desiccation (McWhorter and Hartwig 1965) or freezing (McWhorter 1971).McWhorter and Hartwig (1965) reported that frequent disking operations conducted during the dry summer months usually provide effective control of S. halepense in the southern United States.However, rainfall or irrigation immediately following disking could significantly reduce S. halepense control (Travlos et al. 2019).In this study, tillage was done only once or twice during the mild fall months in the College Station location, and temperatures (September averages: 33 C high and 21 C low; October averages: 27 C high and 15 C low) were not sufficient to effectively desiccate or freeze the rhizomes.Moreover, soil moisture was not a limitation during the fall season (data not shown), which also favored rhizome survival following disking.
It was also evident that flooding is an effective strategy for S. halepense control, but the timing of flooding is critical (Figures 4  and 5C; Table 2).Among the flooding treatments, disking at the initiation of the experiment þ flush irrigation þ flooding after emergence for 14 d was the most effective treatment that consistently reduced S. halepense density, rhizome biomass, rhizome node number, and rhizome length at the termination of the study in spring of 2021, whereas disking at the initiation of the experiment þ disking when S. halepense had regrown to a 30-to 38-cm height þ flooding at first frost (plants were 20-to 30-cm tall, going into the dormant stage) was the least effective flooding treatment.
Flush irrigation following disking likely promoted sprouting of S. halepense, which may have increased the susceptibility of S. halepense to flooding through breaking of rhizome dormancy and through a reduction in the carbohydrate reserves.Although tolerance to flooding is species specific (Barclay and Crawford 1982), plants with higher carbohydrate reserves tend to withstand anaerobic conditions (i.e., flooding) for longer periods, because they can use anaerobic fermentation to maintain metabolic functions for some time (Raju 2007).The low effectiveness of flooding implemented after frost could be associated with the induction of rhizome dormancy in cold temperatures, which in turn may have reduced the physiological response to flooding.Rhizomes of some plants adapted to waterlogged soils can survive all winter under complete anoxia (Crawford 2003), indicating that dormancy and cold temperatures can reduce metabolism enough to prevent tissue death.In previous research in growth chambers, McWhorter (1972b) reported that soil and water temperatures play an important role when flooding is used for S. halepense control.Submerging rhizomes for 8 d in water at 40 C showed excellent control of rhizomes, whereas only a few rhizomes were killed in water at 10 or 20 C during the same period.In the field conditions, however, 2 and 4 wk of continuous flooding were necessary to kill rhizomes in water at 40 and 30 C, respectively; in water at 15 C, most rhizomes survived across 10 wk.Therefore, flooding for only 2 wk in the winter may not have been enough to control S. halepense rhizomes.Flooding for an extended period may be expensive.On the other hand, flooding in the summer months can provide rapid control of S. halepense.A study conducted in Arkansas reported 76% to 85% control of S. halepense when cotton was rotated with flooded rice (Oryza sativa L.) (Frans et al. 1991).In southeast Texas, this can be accomplished during late July to early September after the harvest of cash crops such as corn and grain sorghum.However, irrigation water availability and cost can be significant limitations for this practice.It is also important to notice that flooding can shift the dominance of the weed population from perennial S. halepense to annual weeds (Figure 5C).
Mowing significantly reduced S. halepense density, aboveground biomass, rhizome biomass, and total linear length, but not node number by the end of the experiment in 2021.However, according to our results, repeated mowing is necessary for good control; the impact of mowing was nonsignificant following the first mowing (fall 2019), at measurements taken in spring 2020 or fall 2020.Repeated mowing for at least 2 yr may be required to adequately control rhizomatous S. halepense.Our findings corroborate those of Horowitz (1972b), who also observed reductions in aboveground biomass and rhizome growth of S. halepense after 4 mo of repeated mowing.Mowing causes plants to allocate more resources to shoot growth rather than rhizome expansion, so repeated mowing at regular intervals can prevent rhizome growth and spread.Mowing can also reduce the competitiveness of perennial weeds, create openings in the canopy, and favor other weed species that are more tolerant to mowing (Miller 2016).
The acetic acid application was ineffective for the control of S. halepense.Previous research that evaluated acetic acid  application on various broadleaf and grass weeds has found that weed control increased with increasing concentrations, but efficacy decreased as plants matured (Moran and Greenberg 2008;Webber et al. 2018).Studies have also suggested that acetic acid is generally more effective on small annual broadleaf weeds than perennial grass weeds (Abouziena et al. 2009;Domenghini 2020;Evans et al. 2011;Webber et al. 2018).Perennial plants have thicker leaves and more protective structures around meristems than annual broadleaf plants (Aguirre et al. 2020).This is true for S. halepense, which also has underground rhizomes that are not affected by contact herbicides.Abouziena et al. (2009) noted that 30% acetic acid applied at 188 L ha −1 provided 95% control of seedling S. halepense when the plants were 4-to 7-cm tall, but control declined to 40% when the plants were at 8 to 12 cm.In the present study, acetic acid was applied at 30% concentration on 20-to 30-cm-tall rhizomatous S. halepense.Acetic acid could be an effective treatment against young S. halepense seedlings, but it is not effective on rhizomatous plants.
Black tarping resulted in one of the lowest aboveground S. halepense density and rhizome growth (Figure 5A).Tarping can physically restrict weed growth and development (Lalitha et al. 2010;Zhang et al. 2021).It may also control S. halepense through the solarization effect (Law et al. 2008).Tarping can raise surface soil temperatures to over 60 C in the summer (Candido et al. 2012) and 45 C in the fall (Horowitz et al. 1983).Solarization has been used to control several weed species in vegetable production systems, including perennials such as nutsedges (Cyperus spp.) (Chase et al. 1998;Ricci et al. 1999).Elmore et al. (1993) successfully controlled S. halepense and bermudagrass [Cynodon dactylon (L.) Pers.] rhizomes with clear plastic tarps.Law et al. (2008) reported that solarization using clear plastic with or without prior tillage decreased S. halepense density by more than 56% compared with the bare ground.In this study, we observed that S. halepense rhizomes were desiccated underneath the tarping, which might be attributed to soil solarization (Figure 5B).
In summary, disking alone-once or twice-or acetic acid application for two continuous seasons failed to control S. halepense effectively.Disking should be done multiple times during the hot, dry season, as McWhorter and Hartwig (1965) recommended.Acetic acid shows potential for annual broadleaf control, but it is not effective against perennial grasses.Repeated mowing for 2 yr effectively controlled S. halepense by reducing S. halepense density, aboveground biomass, rhizome biomass, and length, but not node number, indicating that the plants were forced to grow shoots at the expense of rhizomes.However, repeated mowing could favor other problematic annual grass species (Miller 2016).Disking followed by flooding for 14 d is an effective approach for managing S. halepense.Flooding timing is critical, and flooding during the cold season failed to provide adequate control of S. halepense.Flooding is only possible in flat fields and requires structures like berms to keep the water in place.Further, some regions may not have enough water to maintain continuous flood for longer periods, and this method can be very expensive.When flooding is not feasible, the alternative approach of installing tarping can be an excellent option for small areas or high-value crops.Overall, results indicate that well-timed nonchemical management practices implemented during the winter fallow such as tarping and flooding can be very effective in reducing S. halepense densities.It is important to recognize that this study was conducted over a 2-yr period to understand the cumulative effects of the treatments implemented during two consecutive winterfallow seasons.However, due to the extensive effort and resources required to conduct the experiment, it was not repeated in either space or time.As a result, we can infer that the results presented here provide insights into the effect of various nonchemical treatments on S. halepense population growth only under these specific conditions.
Future research should evaluate the effectiveness of a program approach that integrates the different tactics that showed promise in this research.Further, experiments should be done to determine whether approaching ineffective treatments in different ways can improve S. halepense suppression (e.g., disking followed by rhizome collection and removal).However, tillage is not a preferable option due to soil health concerns and would be Weed Science undesirable in no-till organic systems.Thus, more emphasis on conservation practices for S. halepense control is imperative.Winter cover crops such as cereal rye (Secale cereale L.) may be another useful tool for S. halepense control, but this was not included in the current research due to establishment issues during the first year.Future research should investigate cover crops, improving cash crop competitiveness, and other nonchemical tools not included in this research and integrate tactics that can be implemented during the summer cropping season/summer fallow.

Figure 1 .
Figure 1.(A) Initial Sorghum halepense densities at the site (September 2019) and (B) an aerial view of the experimental site during the spring regrowth (April 2020).

Figure 2 .
Figure 2. (A) Implementation of flooding treatments and (B) installation of 6-mil black tarp.

a
at the initiation of the experiment þ disking after the S. halepense had regrown to a 30-to 38-cm height 35 ± 2 a-c 511 ± 192 ab 468 ± 30 a-c T4: Disking at the initiation of the experiment þ flooding for 14 d 7.4 ± 2.4 cd 50 ± 18 bc 83 ± 23 cd T5: Disking at the initiation of the experiment þ flush irrigation þ flooding for 14 d when S. halepense height was 2at the initiation of the experiment þ disking when S. halepense had regrown to a 30-to 38-cm height þ flooding at first frost 29 Values followed by the same letter are not significantly different, based on Tukey's honestly significant difference (HSD) test at the 0.05 probability level.b Total linear length of all rhizomes in a square meter (m 2 ) area.280 Camargo Silva et al.: Organic Sorghum halepense control

Figure 5 .
Figure 5. Photographs of various nonchemical management treatments: (A) impact of tarping on Sorghum halepense densities, (B) desiccation of S. halepense rhizomes underneath the tarping, and (C) impact of flooding on S. halepense densities.Photos were taken in spring 2021.

Table 1 .
Dates of field operations and data collection 278Camargo Silva et al.: Organic Sorghum halepense control

Table 2 .
Sorghum halepense rhizome biomass, nodes, and lengths under different nonchemical management treatments in College Station, TX. a