Plant Weeds

A three-factor randomized block design was used in this experiment. Each treatment had a plot dimension of 22.99 m² (6.13 m×3.75 m), with three replicates for each treatment. The tested rice cultivars, T-43 (V1, a drought-resistant high-yield variety) and Liangxiang-3 (V2, a drought-sensitive variety), are frequently planted in Xinjiang. The seeds of T-43 and Liangxiang-3 were provided by the Institute of Agricultural Science, Tianye Company (Xinjiang, China). The tested rice cultivars were cultivated with two water treatments (W₁, limited drip irrigation at 10,200 m³·hm^-2, which was 85% of the current drip irrigation rate commonly used in rice production in Xinjiang based on preliminary experimental results and W₂, deficit drip irrigation at 8670 m³·hm^-2, which was 85% of the drip irrigation rate used in W₁ during the entire growth period) and three nitrogen fertilization modes with the same total nitrogen application amounts (pure N: 300 kg·hm^-2) and different ratios of seedling fertilizer:tillering fertilizer:panicle fertilizer:grain fertilizer (N₁: 30%:50%:13%:7%; N₂: 20%:40%:30%:10%; and N₃: 10%:30%:40%:20%) (see Table 1 for details). P₂O₅ and K₂O fertilizer were applied with water, and the application rates were 150 kg·hm^-2 and 135 kg·hm^-2, respectively. Fifty percent of the applied P₂O₅ and K₂O was used as seedling fertilizer, and the other 50% was used as panicle fertilizer. Urea (N, 46%) and potassium dihydrogen phosphate (P₂O₅, 52.1%; K₂O, 34.6%) were used as the nitrogen fertilizer and the P₂O₅ and K₂O fertilizer, respectively.
The experiment used a planting pattern of one film, two tubes, and eight rows. The sowing width was 1.65 m, the plant spacing was 10 cm, and the row spacing was 10 cm + 26 cm + 10 cm + 26 cm + 10 cm + 47 cm, as shown in Figure 2. The drip irrigation tape placement, plastic film mulching, spot seeding, and covering of the seeds with soil were all completed at one time. During the growth period, precision management was performed to control pests and weeds in a timely manner. The seeds were sown on May 1, 2020, and May 4, 2021. After being sown, the seeds were drip irrigated with water, and the seedlings were released from the plastic film after emergence. A total of 6 to 8 seedlings were preserved in each planting hole, and the rice plants were harvested on September 30 in both years. The irrigation frequency was once every 3 days before and once every 2 days after jointing, and irrigation was stopped until 15 days before maturity. The drip irrigation tape used in this study was single-wing labyrinth drip irrigation tape produced by Xinjiang Tianye Co., Ltd. (Xinjiang, China) with an emitter spacing of 30 cm and a flow rate of 1.8 L·h^-1. The plots were spaced 1 m apart to prevent lateral water seepage between plots. A water meter (measurement accuracy, 0.001 m³) was used to measure the amount of irrigation water applied through the drip irrigation system. Other field management measures were performed using drip irrigation methods commonly used under plastic film mulching in high-yielding rice fields.

In 2016, initial crosses were carried out by bagging inflorescences of switchgrass ramets in a greenhouse. The majority of crosses occurred between a diverse group of southern genotypes (n = 57) from lowland populations and two lowland genotypes that showed strong winter survivorship as multiple clonal replicates over 6 winters near Arlington, WI. There were also a limited number of crosses between southern genotypes which had not been evaluated for winter survivorship. The winter tolerant genotypes are referred to as Tolerant 1 and Tolerant 2, and they originated from an unknown population originally collected in North Carolina, South Carolina or northern Florida (Timothy DH, personal communication). Collectively, the individuals used for initial crosses will be referred to as Founders. Crosses resulted in 2,058 individuals unevenly distributed across 29 unique F₁ families. The number of individuals per family was the result of variable seed quantity and viability. During the following year, a set of pseudo-F₂ families were generated by crossing randomly selected siblings within F₁ families. This resulted in 1,039 pseudo-F₂ individuals unevenly distributed among 20 full-sib pseudo-F₂ families. Some pseudo-F₂ families were generated from pairs of siblings within an F₁ family, so only 10 F₁ families were represented in the pseudo-F₂ families.
Among controlled greenhouse crosses, the success rate for initial crosses among Founder individuals was 71%, with success defined as resulting in at least one progeny seedling from a parent (mean 36 seedlings per successful cross). Within F₁ sibling matings, used to generate pseudo-F₂ families, the success rate was 20%, but with a mean of 74 seedlings generated per successful cross parent.
All Founder individuals and F₁ parents of pseudo-F₂ families were maintained in a greenhouse and divided into vegetative replicates by dividing crowns. In July 2018, a completely randomized spaced plant nursery was planted with 195 genotypes. The spaced plants were genotypes maintained in 12-plant rows with 0.7 m between and 0.7 m within rows. Weeds were controlled between individuals genotype crowns using roto-tilling and occasional hand weeding. The nursery contained a minimum of two vegetative replicates per individual. In 2019, vegetative replicates reserved from Founder individuals and F₁ parents of pseudo-F₂ families were used to replace individuals that were lost to winterkill in the spaced plant nursery during the first winter.
In addition, an unreplicated, stratified by genotype spaced row nursery (unique genotypes planted with 0.7 m between rows and 0.3 m within rows) was established of the F₁ families and pseudo-F₂ families in the spring of 2018. Each row contained 10 unique genotypes. In the summer of 2018, heavy rain and standing water in sections of the nursery and resulted in uneven and poor plant vigor. To account for establishment damage that was unrelated to winter survival, fall vigor ratings were made on a scale of 0 to 5 during September in 2018 and 2019. Fall vigor was then used as a covariate for the subsequent spring vigor scores. A fall vigor score of 5 indicated a healthy switchgrass plant and zero indicated a deceased plant.
Winter survivorship scores and heading date was measured for each individual in both nurseries during 2019, 2020, and 2021 (spring vigor only). Spring vigor was recorded using a scale from 0–20, with 20 indicating no visible damage and 0 indicating mortality. Heading date was recorded as the date in which panicles were observed on at least 50% of an individual's tillers.

In May 2017, three plants per species and plot (if possible) were selected and plant height (cm) was measured as the stretched length of three vegetative shoots per individual. First, the heights of the three shoots per individual were averaged and then the mean values of the three individuals per plot. After that, bulk samples of 10–15 fully developed leaves were collected from the same individuals and shoots (one to three leaves per shoot). Leaves were stored in sealed plastic bags in a cooling box for transport to the laboratory, where leaf area (mm²_leaf) was measured with a leaf area meter (LI‐3000C Area Meter equipped with LI3050C transparent belt conveyor accessory; LI‐COR). Then, leaf samples were dried for 48 h at 70°C, weighed, and specific leaf area (SLA; mm²_leaf mg⁻¹_dw) was calculated as the ratio between total leaf area and total leaf mass per plot and species. Dry leaf samples were ground to a fine powder with a mixer mill (MM2000, Retsch). Approximately 10 mg of the milled material was then used to determine leaf nitrogen concentration (mg N g⁻¹_leaf) with an elemental analyzer (Vario EL cube, Elementar Analysensysteme GmbH). Leaf phosphorus (mg P g⁻¹_leaf) and leaf potassium concentration (mg K g⁻¹_leaf) were measured using an inductively‐coupled plasma optical emission spectrometer (Thermo Scientific™ iCAP™ 7400 ICP‐OES Duo). Therefore, milled leaf samples (250 mg) were first treated in a Mars 6 microwave closed system (CEM GmbH) for acid digestion (with 5 mL of HNO₃ and 0.5 mL of H₂O₂) and then the diluted acid extracts were analyzed with the ICP‐OES to measure P and K.
For the determination of root colonization by arbuscular mycorrhizal fungi (AMF), we collected roots of the selected plant individuals by taking soil cores (10 cm depth, 5 cm diameter), which contained the root crown and attached roots of the plants. Soil was roughly removed and roots per plot and species were stored in plastic bags. In the laboratory, roots were cleaned by rinsing off the remaining soil with tap water, and then the material was stored in 70% ethanol until further processing. For the determination of AMF colonization, roots were first rinsed with tap water to remove ethanol, and then, a subsample of ~20 g of finer roots was purified by heating in 10% potassium hydroxide solution at 80°C for 30–90 min (heating times varied depending on plant species). After this, roots were heated for 5 min at 80°C in an ink–vinegar solution (5% black ink: Parker S0037460 Quink Black; 95% vinegar: white household vinegar, 5% acetic acid) to stain AMF following Vierheilig et al. (1998 (link)). After staining, roots were rinsed several times with and stored in a water‐vinegar mixture to remove excess stain. Finally, AMF colonization was scored under the microscope (200x magnification) using the line‐intersect method for 100 intersects (McGonigle et al., 1990 (link)).
For the determination of community‐level root traits, we took two soil cores (10 cm depth, 5 cm diameter) per plot in June 2017 in the inner center, i.e., with a distance of at least 30 cm from the plot edge. Soil cores were pooled per plot and stored in a freezer until further analysis (−20°C). Later, soil cores were defrosted, and roots were cleaned with tap water. Then, root samples per plot were scanned with a flatbed scanner at 800 dpi (Epson Expression 10000 XL scanner, Regent Instruments), and root length was measured with an image analysis software (WinRHIZO; Regent Instruments), followed by drying (at 70°C for 48 h) and weighing. Specific root length (SRL) was calculated as the ratio between root length and root dry mass (m_root g_root⁻¹), and root length density (RLD) as the ratio of root length to volume of the soil cores (cm_root cm_soil⁻³).
Aboveground biomass was harvested block‐wise on each plot from 29 May to 5 June 2017. A sample area of 0.2 × 0.5 m was chosen in the inner center of the plots, and plants were cut 3 cm above ground. Biomass samples were sorted to sown plant species, weeds, and dead plant material, dried at 70°C for 48 h, and weighed. Total aboveground biomass of the sown plant species per plot was extrapolated to one square meter (g m⁻²) as a measure of community biomass production.