Introduction
Fall panicum is a C4 native grass found throughout the United States in several cropping systems and noncultivated areas (Bryson and DeFelice Reference Bryson and DeFelice2009; Odero and Sellers Reference Odero and Sellers2022; Ohsugi and Murata Reference Ohsugi and Murata1986). Seedlings of fall panicum emerge throughout the spring and summer in cultivated fields and postagricultural successional lands (Vengris and Damon Reference Vengris and Damon1976). In the Florida rice (Oryza sativa L.) production system in the Everglades Agricultural Area (EAA) in southern Florida, fall panicum is the most prevalent weed species, emerging year-round in the subtropical climate of the region (Cherry and Bennett Reference Cherry and Bennett2005; Odero and VanWeelden Reference Odero and VanWeelden2018). The EAA is dominated by organic soils (Histosols) in an area covering approximately 280,000 ha (Daroub et al. Reference Daroub, Horn, Lang and Diaz2011). Prevalence of fall panicum in the region is attributed to limited control options in the cropping system, which involves rotation of sugarcane (Saccharum spp. hybrids), the main crop, with grass crops including rice and sweet corn (Zea mays L. var. saccharata). The ability of fall panicum to continuously replenish the soil seed bank through prolific seed production and its persistence in the soil for several years (Alex Reference Alex1980; Burnside et al. Reference Burnside, Fenster, Evetts and Mumm1981) have also enabled it to dominate the cropping system in the region.
Rice has been successfully integrated as a rotational crop with sugarcane in the EAA during sugarcane’s fallow period. Over 20,000 ha of fallow sugarcane fields in the EAA are available for rice production every year (Bhadha et al. Reference Bhadha, Trotta and VanWeelden2019). In 2021, approximately 10,000 ha of rice were planted in the region (M VanWeelden, personal communication). Rice in the EAA is occasionally used as a cover crop before planting sugarcane, sweet corn, or leafy vegetables. Straw and stubble from the rice cover crop provides benefits such as improvement of the soil texture, tilth, and drainage (Schueneman et al. Reference Schueneman, Rainbolt and Gilbert2008). Cultivation of rice on fields in the EAA that would otherwise be fallow has been shown to result in increased yield of the subsequent sugarcane crop (Alvarez and Snyder Reference Alvarez and Snyder1984). The rice cover crop in the region is occasionally grown under nonflooded conditions, which results in limited soil moisture available for growth particularly for early-planted rice. Rice in the EAA is usually planted from February or early March and harvested until September or early October. The early planting period coincides with the dry season in the EAA that occurs between November and May (Lang et al. Reference Lang, Oladeji, Josan and Daroub2010). Fall panicum emergence, growth, and development also occurs during the dry season when rice planting commences in the EAA. The limited availability of soil moisture under such dry conditions not only influences growth and development of crops, but also that of weed species (Fernandez-Quinantilla et al. Reference Fernandez-Quinantilla, Andujar and Appleby1990). Biotypes of weed species such as African turnip (Sisymbrium thellungii O.E. Schulz) can differ in their response to limited soil moisture by exhibiting differences in physiological behavior (Mahajan et al. Reference Mahajan, George-Jaeggli, Walsh and Chauhan2018).
Weeds compete with rice for nutrients, moisture, and light throughout the growing season, resulting in yield reduction (Odero and VanWeelden Reference Odero and VanWeelden2018). In addition, weeds are alternate hosts for pathogens and insect pests of rice (Cherry and Odero Reference Cherry and Odero2021; Wisler and Norris Reference Wisler and Norris2005; Yadav and Thrimurty Reference Yadav and Thrimurty2006;). The most negative effect of weeds on crop growth and yield is competition for limited environmental resources required for growth (Patterson Reference Patterson1995). Crop growth and productivity are often constrained by environmental factors like limited water availability (Kramer Reference Kramer, Turner and Kramer1980). Such stressful environmental factors influence weed–crop interactions by changing the physical environment for plant growth and can adversely affect the growth of weeds and the level of their control (Patterson Reference Patterson1995). Weed–crop competition for water reduces the quantity of water available in the soil for crop growth and contributes to crop water stress (Patterson Reference Patterson and Duke1985; Zimdahl Reference Zimdahl2004). Water use efficiency, transpiration, and response to water stress differs significantly among weed species and crops, consequently affecting the weed–crop competition process (Geddes et al. Reference Geddes, Scott and Oliver1979; Patterson Reference Patterson1986; Patterson and Flint Reference Patterson and Flint1983). For example, soybean [Glycine max (L.) Merr.] growth and yield reduction from jimsonweed (Datura stramonium L.) competition is highest when soil moisture is limited. The growth of smooth pigweed (Amaranthus hybridus L.) in competition with cotton (Gossypium hirsutum L.) is less affected by low moisture availability compared with cotton (Stuart et al. Reference Stuart, Harrison, Abernathy, Krieg and Wendt1984). Water stress can also influence the duration of the critical weed-free period for different crops (Patterson Reference Patterson1995).
Adaptive mechanisms developed by plants to ameliorate the impact of water stress differs among species and depends on soil type, nutrients, climatic patterns, and the extent of water stress (Akinci and Lösel Reference Akinci, Lösel, Rahman and Hasegawa2011). Physiological, anatomical, and morphological changes at the leaf and whole-plant levels, which plants develop to tolerate water stress, can impose costs on plant growth and result in reduced growth and development (Pugnaire et al. Reference Pugnaire, Serrano, Pardos and Pessarakli1999). In common waterhemp (Amaranthus rudis Sauer), water stress reduced vegetative development and seed production (Sarangi et al. Reference Sarangi, Irmak, Lindquist, Knezevic and Jhala2015). Furthermore, under severe water stress conditions, common waterhemp did not produce any seeds, because the plants did not survive more than 30 d (Sarangi et al. Reference Sarangi, Irmak, Lindquist, Knezevic and Jhala2015). The growth and seed production of jungle rice [Echinochloa colona (L.) Link] decreased at soil moisture content of 25% field capacity or below (Chauhan and Johnson Reference Chauhan and Johnson2010). For itchgrass [Rottboellia cochinchinensis (Lour.) W.D. Clayton], although growth and seed production decreased with increasing water stress, both were lower at 100% than at 75% field capacity (Chauhan Reference Chauhan2013). Growth and reproductive behavior of wild oat (Avena fatua L.) and sterile oat (A. ludoviciana L.) decreased with increasing water stress (Sahil et al. Reference Sahil, Loura, Raymont and Chauhan2020). Bengal dayflower (Commelina benghalensis L.), spiny amaranth (Amaranthus spinosus L.), and Chinese sprangletop [Leptochloa chinensis (L.) Nees] have also been reported to have the ability to produce shoot biomass and reproduce under water stress (Chauhan and Abugho Reference Chauhan and Abugho2013; Webster and Grey Reference Webster and Grey2008). As a C4 plant, fall panicum has a carbon fixation pathway that confers on it the ability to be more tolerant to water stress (Ohsugi and Murata Reference Ohsugi and Murata1986). Therefore, it is important to determine its growth and reproductive capacity under soil water stress conditions. Currently, there is limited information on fall panicum growth and reproduction under soil water stress conditions. Therefore, the objective of this study was to evaluate the effect of different soil moisture levels on growth and reproductive capacity of fall panicum.
Materials and Methods
Greenhouse studies were conducted at the Everglades Research and Education Center (EREC) in Belle Glade, FL, between December 2020 and May 2021 to determine fall panicum growth and reproductive behavior under different soil water contents. Two populations of fall panicum seeds were collected at maturity from a sugarcane field at the EREC (26.6642° N, 80.6375° W) and at the Florida Sugarcane Growers Cooperative (FSGC) Glades Farm (26.7073° N, 80.5464° W) in the summer of 2020 and stored separately in the dark at 2 C before use. The history of the fields involved rotation of sugarcane, the main crop, with rice, sweet corn, or leafy vegetables. The seeds were planted in growing trays measuring 53 by 28 cm filled with a commercial potting medium (Fafard®; Sun Gro Horticulture, Agawam, MA) on December 20, 2020, and February 5, 2021, for the first and second experimental runs, respectively, prior to transplanting the plants into pots filled with organic soil. A slow-release 14-14-14 fertilizer (Osmocote®; The Scotts Company, Marysville, OH) at 140 g N kg–1, 61 g P kg–1, and 116 g K kg–1 was mixed with the potting medium, and the plants were kept in a greenhouse with a maximum temperature of 30 C under natural light for the entire study. Fall panicum was transplanted into round free-draining 7.6-L (16 cm diam and 17 cm height) pots filled with organic soil collected from a sugarcane field at the EREC. The soil was Dania muck (Euic, hyperthermic, shallow Lithic Haplosaprists) with 7.3 pH and 74% organic matter. Soil pH and organic matter content were determined using the method described by Fernandez et al. (Reference Fernandez, Odero, MacDonald, Ferrell, Sellers and Wilson2019). The soil was air-dried and passed through a 13-mm sieve to remove debris prior to use. No fertilizer was added to the organic soil.
The pot (soil) water content, hereafter referred to as soil water content (SWC) of the organic soil, was assessed using previously described methods (Sarangi et al. Reference Sarangi, Irmak, Lindquist, Knezevic and Jhala2015; Steadman et al. Reference Steadman, Ellery, Chapman, Moore and Turner2004). Each pot was filled with 3.18 kg of dried soil and watered until the soil was saturated. The pots were covered with aluminum foil to prevent evaporation and allowed to drain freely for 36 h. After 36 h, the pots were reweighed to determine the SWC using Equation 1:
$$SWC = \left[ {\left( {{W_w}-{W_d}} \right)/D} \right]$$
where W w is the wet weight of the soil in the pot, W d is the dry weight of the soil in the pot, D is the density of the water (Sarangi et al. Reference Sarangi, Irmak, Lindquist, Knezevic and Jhala2015).
A completely randomized design with a five-by-two factorial treatment arrangement and five replications was used to set up the experiment. The first factor consisted of five SWC levels (100%, 75%, 50%, 25%, and 12.5%), whereas the second factor consisted of two fall panicum populations (EREC and FSGC biotypes). The 75%, 50%, 25%, and 12.5% SWC levels were estimated as a fraction of the 100% SWC. A 3-d pot watering interval for fall panicum to maintain the desired SWC based on the weight of each pot was used in the experiment based on previous work by Negrisoli (Reference Negrisoli2019).
One fall panicum seedling 2 cm in height was transplanted from the growing trays into a round, free-draining 7.6-L pot filled with organic soil. Fall panicum was allowed to grow for 7 d after transplanting and watered sparingly to ensure the plants overcame transplanting shock before commencing the SWC treatments. SWC level treatments were initiated on January 5, 2021, for the first experimental run and February 22, 2021, for the second experimental run.
Fall panicum height, number of leaves per plant, and number of tillers per plant were recorded at 7-d intervals for 56 d after treatments (DAT) were initiated. At the end of the study, fall panicum panicle branches from each pot were quantified to determine its reproductive capacity. Panicle branches were used to estimate the reproductive capacity of fall panicum, because the tendency of seeds to shatter very quickly after ripening complicates estimation of the number of seeds produced (Govinthasamy and Cavers Reference Govinthasamy and Cavers1995). The plants were separated into leaves and stems at 56 DAT, and total leaf area was measured using a LI-COR leaf area meter (LI-3000C Portable Area Meter; LI-COR Biosciences, Lincoln, NE). The leaves and stems (shoot biomass) of each fall panicum plant were bagged separately. The roots (root biomass) of the plants were then collected after removing soil particles with a gentle wash using water. The shoot and root biomass were dried at 60 C for 72 h and weighed to determine dry biomass. Root-to-shoot ratio was calculated as the ratio of root biomass to biomass of the shoot.
Statistical Analysis
All data were checked for normality and homogeneity of variance using Shapiro-Wilks and Levene tests, respectively, in R (R Core Team 2021) before analysis and transformed when necessary. Data were subjected to ANOVA using the lme4 package (Lenth Reference Lenth2021) of the R statistical language version 4.1.0 (https://cran.r-project.org/bin/windows/base/). SWC, fall panicum biotype, and their interactions were considered fixed effects, whereas experimental run and replication (nested within experimental run) were considered random effects. Linear regression models were performed on plant height and number of leaves per plant using the nlme package of R (Pinheiro et al. Reference Pinheiro, Bates, DebRoy and Sarkar2021):
$$Y = ax + b{x^2} + c$$
$$Y = ax + c$$
where Y is the response (plant height or number of leaves per plant) at time x, x is the time expressed as days after SWC treatment initiation, a and b are regression coefficients, and c is the plant height or number of leaves per plant at initiation of the SWC treatments at x = 0. A lack-of-fit test at the 95% level was conducted to determine whether the models (Equations 2 and 3) appropriately fit the data. Nonlinear regression analysis was performed on the number of tillers per plant using the drc package of R (Ritz and Streibig Reference Ritz and Streibig2005, Reference Ritz and Streibig2016). The data were analyzed using a four-parameter log-logistic model:
$$Y = c + \left\{ {d-c/1 + {\rm exp}\left[ {b\ \left( {{\rm log}\ x-{\rm log}\ e} \right)} \right]} \right\}$$
where Y is the number of tillers per plant at time x, x is the time expressed as days after SWC treatment initiation, b is the slope at the inflection point, c is the lower limit, d is the upper limit or asymptote, and e is the number of days after SWC treatment initiation where the inflection point occurs. The model (Equation 4) was selected based on Akaike’s information criterion using of the qpcR package of R (Spiess Reference Spiess2018). A lack-of-fit test at the 95% level comparing the model (Equation 4) to ANOVA was conducted to determine whether the model appropriately fit the data (Ritz and Streibig Reference Ritz and Streibig2005). Root-mean-square error and adjusted R2 were calculated using the qpcR package of R to test for goodness of fit of the nonlinear and linear models, respectively. Estimated marginal means were calculated for total leaf area, shoot biomass, root biomass, number of panicles per plant, and root-to-shoot ratio, and the post-hoc Tukey test performed for all pairwise comparisons (P < 0.05) using the emmeans package of R (Bates et al. Reference Bates, Maechler, Bolker and Walker2021).
Results and Discussion
The main effect of SWC treatment over time was significant for fall panicum height, number of leaves per plant, number of tillers per plant over the duration of the study; however, the main effect of fall panicum population and the interactions of the main effects were not significant (Table 1). The two populations of fall panicum used in this study exhibited similar growth in response to soil water stress despite originating from different locations, most likely because of their adaptation to a similar cropping system that involves rotation of sugarcane with rice, sweet corn, or leafy vegetables.
Table 1. Significance of main effects of pot soil water content (SWC), fall panicum biotype (FPB), and their interaction over time for fall panicum height, number of leaves per plant, and number of tillers per plant in greenhouse experiments in Belle Glade, FL in 2021. a
a Asterisks denote significant at ***P < 0.001.
Linear regression models provided the best fit for fall panicum height and number of leaves per plant produced in response to different SWC over time (Table 2). Fall panicum height and number of leaves per plant decreased as SWC levels decreased at all evaluation timings despite continuous increase in both growth parameters from 7 to 56 DAT initiation (Figure 1). The maximum height of fall panicum was estimated to be 125 cm at 100% SWC, whereas plant height at 75% and 50% SWC were estimated to be 95 and 85 cm, respectively. Fall panicum height declined by 65% and 50% at 12.5% and 25% SWC, respectively, relative to the maximum height achieved at 100% SWC. These results show that increase in fall panicum height, which is a function of stem elongation (Sachs Reference Sachs1965), was sensitive to soil water stress. Plant height has been shown to be sensitive to soil water stress in several species including Miscanthus × giganteus (Greef & Deuter ex Hodkinson & Renvoize), common waterhemp, jungle rice, and itchgrass (Chauhan Reference Chauhan2013; Chauhan and Jonson Reference Chauhan and Johnson2010; Ings et al. Reference Ings, Mur, Robson and Bosch2013; Sarangi et al. Reference Sarangi, Irmak, Lindquist, Knezevic and Jhala2015). In a similar study, the height of common waterhemp decreased by 43% at 25% pot SWC and 71% at 12.5% pot SWC compared with plants at 100% pot SWC (Sarangi et al. Reference Sarangi, Irmak, Lindquist, Knezevic and Jhala2015). The decrease in fall panicum height under soil water stress was probably attributed to reduced cell elongation associated with loss of cell turgor when water availability required for cell division and growth is limited (Farooq et al. Reference Farooq, Wahid, Kobayashi, Fujitha and Basra2009; Tátrai et al. Reference Tátrai, Sanoubar, Pluhár, Mancarella, Orsini and Gianquinto2016).
Table 2. Model parameters and SEs in parentheses for linear models for fall panicum plant height, number of leaves per plant, and the goodness of fit (adjusted R2) in response to different pot soil water content (SWC) levels over time in greenhouse experiments combined over two biotypes and experimental runs in Belle Glade, FL in 2021. a
a Equations 2 and 3: Y = ax + bx 2 + c and Y = ax + c. Y is the response (plant height or number of leaves per plant) at time x, x is the time expressed as days after SWC treatment initiation, a and b are regression coefficients, and c is the plant height or number of leaves per plant at initiation of the SWC treatments at x = 0.
b ±SE is given in parentheses.
c Denotes Equation 2 was used.
d Denotes Equation 3 was used.
Figure 1. Fall panicum (A) height, (B) number of leaves, and (C) number of tillers in response to pot soil water content over time.
Like fall panicum height, leaf production decreased as SWC levels decreased over time (Figure 1B). Fall panicum produced the highest number of leaves (58 leaves per plant) at 100% SWC. The maximum leaves produced at 75% and 50% SWC were 53 and 46 leaves per plant, respectively. A similar finding was reported for prickly lettuce (Lactuca serriola L.), where leaf production was highest at 100% SWC compared with decreasing levels of SWC (Chadha et al. Reference Chadha, Florentine, Chauhan, Long and Jayasundera2019). In contrast, Chauhan (Reference Chauhan2013) reported that the number of leaves produced by itchgrass was lower at 100% of soil field capacity compared with 75% and 50% of soil field capacity.
The log-logistic model provided the best fit to estimate fall panicum tiller production in response to different SWC levels over time. The model estimated that a maximum of approximately nine tillers per plant were produced at 75% and 100% SWC (Table 3). Maximum tiller production at 25% and 12.5% SWC was estimated to be four and five tillers, respectively, which was equivalent to 56% and 46% reduction, respectively, compared with 100% SWC. Based on the log-logistic model, plants at 50% to 100% SWC were able to achieve 50% tiller production (parameter e) within 31 to 43 d compared with 28 d at 25% SWC (Table 3). However, 50% tiller production at the severe soil water stress level (12.5% SWC) was not reached during the duration of the experiment but would have been reached at 68 d. Chauhan (Reference Chauhan2013) reported an exponential growth response of itchgrass tiller number over time at different soil water stress levels. In their study, itchgrass produced 22 tillers per plant at 100% of field capacity. Chinese sprangletop was able to survive and produce significant numbers of tillers even at the lowest soil moisture content of 12.5%, which resulted in 5 tillers per plant compared with 9 to 10 tillers per plant at 25% to 100% soil moisture content (Chauhan and Abugho Reference Chauhan and Abugho2013). Although fall panicum growth parameters (height, leaves per plant, and tillers per plant) decreased significantly at the most severe water stress level (12.5% SWC), no mortality of the plants occurred for the duration of the study. These results show that fall panicum will survive but will not achieve maximum growth under limited soil moisture.
Table 3. Model parameters and SEs in parentheses for the log-logistic model for fall panicum tillers and the goodness of fit (root-mean-square error, RMSE) in response to different pot soil water content (SWC) levels over time in greenhouse experiments combined over two biotypes and experimental runs in Belle Glade, FL in 2021. a
a Equation 4: Y = c + {d – c/1 + exp [b (log x – log e)]}. Y is the number of tillers per plant at time x, x is the time expressed as days after water stress treatment initiation, b is the slope at the inflection point, c is the lower limit, d is the upper limit or asymptote, and e is the number of days after water stress treatment initiation where the inflection point occurs.
b ±SE is given in parentheses.
There was significant effect of SWC on shoot biomass, root biomass, total leaf area, and number of panicle branches per plant at 56 DAT (Table 4). Both shoot (leaves and stem) biomass and root biomass decreased as SWC decreased (Figure 2). Plants under no water stress (100% SWC) had the maximum shoot biomass (23 g plant–1) and root biomass (5 g plant–1). Similar findings were reported for jungle rice, itchgrass, wild oat, sterile oat, and common waterhemp (Chauhan Reference Chauhan2013; Chauhan and Johnson Reference Chauhan and Johnson2010; Sahil et al. Reference Sahil, Loura, Raymont and Chauhan2020; Sarangi et al. Reference Sarangi, Irmak, Lindquist, Knezevic and Jhala2015). Approximately 85% and 68% reduction in shoot biomass was recorded at 12.5% and 25% SWC, respectively, relative to 100% SWC. Plants grown under 100% SWC had more shoot biomass with more biomass partitioning to leaves and stems compared to plants under low SWC levels (Figure 2). Within the same SWC level, shoot biomass was partitioned more to the stems than leaves. In contrast, itchgrass allocated more biomass to leaves compared with stems at high levels of water stress (Chauhan Reference Chauhan2013). Like shoot biomass, more root biomass was produced at 100% SWC compared with 12.5% and 25% SWC, which produced 83% and 66% less biomass, respectively (Figure 2). The root-to-shoot ratio was significantly different between the SWC levels (Table 4). Plants at 12.5% SWC had higher root-to-shoot ratio compared to plants at 100% SWC, indicating that they partitioned more biomass to the roots to facilitate absorption of the limited moisture efficiently. A similar response was observed in prickly lettuce, where the lowest SWC level (25% water holding capacity) produced the highest root-to-shoot ratio (Chadha et al. Reference Chadha, Florentine, Chauhan, Long and Jayasundera2019). The high root-to-shoot ratio under water stress conditions is a morphological adaptation attributed to more biomass partitioning to roots to maximize water uptake (Wang and Taub Reference Wang and Taub2010).
Table 4. Significance of main effects of pot soil water content (SWC), fall panicum biotype (FPB), and their interaction for fall panicum total leaf area, shoot biomass, root biomass, and number of panicle branches per plant at 56 d after SWC treatment initiation in greenhouse experiments in Belle Glade, FL in 2021. a
a Asterisks denote significant at *P<0.05 and ***P < 0.001.
Figure 2. Effect of pot soil water content (SWC) on fall panicum (A) shoot biomass and its partitioning to leaves and stems, (B) root biomass, and (C) root-to-shoot ratio at 56 d after SWC treatment initiation. Means followed by the same letter for either total shoot biomass, stem biomass, leaf biomass, root biomass, or root-to-shoot ratio are not significantly different at P < 0.05. Error bars indicate one standard deviation.
Fall panicum total leaf area significantly increased as SWC content increased at 56 DAT (Figure 3). Plants in the 100% SWC treatment produced the highest leaf area (1,102 cm2 plant–1). There was significant reduction of 23%, 45%, 65%, and 79% in total leaf area at 75%, 50%, 25%, and 12.5% SWC levels, respectively, relative to the 100% SWC level. Similar total leaf area reduction with increasing soil water stress was reported for common waterhemp and prickly lettuce (Chadha et al. Reference Chadha, Florentine, Chauhan, Long and Jayasundera2019; Sarangi et al. Reference Sarangi, Irmak, Lindquist, Knezevic and Jhala2015). Reduced leaf area is an adaptation mechanism of plants under water stress that limits their transpirational surface by inhibiting both cell production and expansion (Alves and Setter Reference Alves and Setter2004). In the present study, reduction in leaf area as soil water stress increased was associated with decrease in leaf production under similar conditions (Figure 3).
Figure 3. Effect of pot soil water content (SWC) on fall panicum (A) total leaf area, and (B) number of panicle branches at 56 d after SWC treatment initiation. Means followed by the same letter for total leaf area or number of tillers are not significantly different at P < 0.05. Error bars indicate one standard deviation.
Production of fall panicum panicle branches, which was a measure of its reproductive capacity, decreased with increasing levels of soil water stress at 56 DAT. The highest number of panicle branches (609 per plant) was produced at 100% SWC (Figure 3). Fall panicum was able to produce panicles at all SWC levels, although the number of panicle branches was significantly lower as severity of soil water stress increased. Plants at 12.5% and 25% SWC produced 82% and 59% fewer panicle branches, respectively, compared with plants at 100% SWC. The inflorescence of fall panicum is a large, open, and wide-spreading panicle consisting of branches that bear spikelets containing glumes and florets that produce seeds that shatter soon after ripening (Odero and Sellers Reference Odero and Sellers2022). The number of spikelets and consequently the number of seeds of grasses that produce panicles is related to the number of panicle branches (AL-Tam et al. Reference AL-Tam, Adam, Anjos Ad, Larmande, Ghesquière, Jouannic and Shahbazkia2013).
This study showed that SWC or soil water stress influences the growth and reproductive capacity of fall panicum. Fall panicum achieved the greatest growth and reproductive capacity at 100% SWC. Although fall panicum did not reach its full growth potential at high soil water stress levels, it was able to survive and develop panicles, showing its ability to adapt to dry conditions associated with early-planted rice in the EAA. Similarly, itchgrass was able to accumulate biomass and produce seeds at 12.5% and 25% SWC (Chauhan Reference Chauhan2013). Plants have several adaptive mechanisms to withstand water stress. Fall panicum survival under water stress was probably due to partitioning more biomass to roots to enhance water uptake and having smaller leaf area to reduce transpirational loss and enhance water use efficiency. The decrease in fall panicum height at lower SWC was probably attributed to decreased cell elongation due to reduction in turgor pressure at low soil moisture conditions (Farooq et al. Reference Farooq, Wahid, Kobayashi, Fujitha and Basra2009; Tátrai et al. Reference Tátrai, Sanoubar, Pluhár, Mancarella, Orsini and Gianquinto2016). Although SWC levels below 50% reduced fall panicum growth and reproductive capacity, it could probably still result in rice yield loss if not efficiently controlled as a result of increase in competition for the limited soil moisture (Scott and Geddes Reference Scott and Geddes1979). Chauhan and Abugho (Reference Chauhan and Abugho2013) reported that rice did not survive at 12.5% and 25% SWC compared with spiny amaranth and Chinese sprangletop, which grew to maturity like the fall panicum growth observed in the present study. Therefore, for early-planted nonflooded cover crop rice in the EAA under dry conditions when the crop is vulnerable to competition from weeds for limited soil moisture, it is important to control the initial fall panicum cohorts that emerge with or prior to rice. Even though the reproductive capacity of fall panicum decreased with SWC level reduction, it was able to produce panicle branches at 25% and 12.5% SWC. Seeds produced from these panicles can replenish the soil seedbank and become sources of re-infestation in subsequent crops. Reducing the weed seedbank is an important proactive long-term weed management strategy to limit weed infestation in cropping systems (Buhler et al. Reference Buhler, Hartzler and Forcella1997). Because of reduced postemergence herbicide uptake by weeds under water stress (Ahmadi et al. Reference Ahmadi, Haderlie and Wicks1980; Dickson et al. Reference Dickson, Andrews, Field and Dickson1990), it is important that future research concentrate on determining the influence of soil water stress levels on rice and fall panicum competition in response to grass herbicides used in rice production.
Acknowledgments
We thank the weed science program staff at the Everglades Research and Education Center for their assistance with the project. This project was supported by the USDA National Institute of Food and Agriculture, Hatch project (FLA-ERC-005755). No conflicts of interest have been declared.
