Decades of this practice with little to no crop rotation, however, have selected for populations of grass species that can escape flooding, among them early watergrass [Echinochloa oryzoides Fritsch], late watergrass [E. oryzicola Vasinger] and bearded sprangletop [Leptochloa fusca Kunth ssp. fascicularis N. Snow] . In addition, season long flooding favors aquatic weeds and promotes the early season growth of algae. Aquatic weeds and algae can shade underwater rice seedlings and delay or reduce rice emergence and early-season vigor, which can negatively impact the stand’s competitiveness with emerged weeds . Although good water management and other cultural practices continue to provide grass weed suppression for many CA rice growers, the vast majority rely on herbicides as their principal means of weed management . California rice has a limited number of available herbicides, due to high costs of development and registration , as well as regulations based on concerns of herbicide drift from aerial applications that may damage neighboring orchards . Similarly, the modes of action of available herbicides are also limited, commercial grow racks and in most cases there is only one herbicide for a given MOA . Repeated yearly applications of the same herbicides for control of specific weeds in many CA rice fields have thus resulted in widespread herbicide resistance in the region.
Herbicide resistance management has become a paramount concern for rice growers, researchers, and the CA rice industry . Efforts to promote herbicide resistance mitigation in CA largely focus on rotation of the limited number of available herbicides, while the cropping strategy itself has remained largely static. Given a robust and fully-amortized infrastructure for water-seeding, any alternative seeding and management techniques for California rice must be economical and provide competitive yields to be appealing to growers . Alternative rice cropping strategies in CA typically only involve modifications to the current water-seeded continuously-flooded system . One strategy used by some growers is the “stale seedbed” method. In this method, rice seedbeds are prepared in the spring, and shallowly irrigated to promote weed germination and emergence. Emerged weeds can then be managed via cultural methods such as cultivation, or via chemical methods such as non-selective herbicides such as glyphosate or paraquat . Afterwards, the fields are flooded and seeded as usual. This method can accomplish a reduction of the general soil seed bank, as well as a reduction of herbicide-resistant weeds in a field . However, high-clay CA fields take several weeks to dry-down enough for field equipment after spring rains, and there is a general perception among CA growers that stale seedbed usage will delay rice planting by too long, shortening the season and potentially depressing yields.
For this reason, stale seedbed implementation is usually limited to fields with severe herbicide resistance problems. Another rice cropping strategy used in other parts of the US in conjunction with a stale seedbed, and which could be adapted to aid herbicide resistance management efforts in California, is drill seeding.Drill seeding of rice is uncommon in California, though it is widely practiced in the US Mid-South. In that region, dry rice seed is drilled to 1.25 – 2 cm, and fields are typically flush-irrigated for the first few weeks as the rice stand develops and herbicides are applied, after which the field is flooded for the remainder of the season . This cropping method discourages aquatic weed species and early-season algal growth , but it can encourage grasses that emerge at the same time as the rice, especially when the crop is sown to the same depth as competitive weeds like barnyardgrass P. Beauv. It is commonly accepted that weed germination will be higher in aerobic soils such as found in drill seeded fields, though germination will generally decrease with increasing seed burial depth . For instance, Benvenuti, Macchia, and Miele found that barnyard grass seedling emergence was 20% or less at burial depths greater than 4 cm, and that mean time to emergence was 10 days or more at these depths. Chauhan and Johnson , however, found that barnyard grass emergence was less than 5% at greater than 4 cm depth, with a mean time to emergence of greater than seven days.
Seeding rice deeper than 4 cm could place the rice below the zone of maximal weed germination, thus delaying stand emergence and allowing the safe use of non-selective herbicides on emerged weeds just prior to rice emergence. This deep drilled modification of the stale seedbed preparation could allow rice growers to plant on time while managing herbicide resistance. Deeper planting of rice is generally not recommended for U.S. Mid-South cultivars, as lower seedling vigor may decrease rice emergence. Koger, Walker, and Krutz found that a 3.8 cm seeding depth reduced the emergence of ‘Lemont’, ‘Cocodrie’, and ‘Wells’ cultivars by 32%, 34%,and 20%, respectively at 21 days after planting . Stand emergence is a principal concern of deep-seeded rice, as reduced or uneven stands typically have lower yields despite increased tillering, and uneven ripening due to nonsynchronous emergence may reduce milling quality . Therefore, rice cultivars with greater vigor would be expected to have more rapid emergence and more even stands than lower-vigor cultivars when sown at greater depths . Elongating seedlings of cereal crops sown below the soil surface consist of a coleoptile, which covers and protects the first leaf during elongation towards the surface, and a mesocotyl, which is the first internode between the coleoptile and the seed . In dry-seeded cereals, mesocotyl elongation tends to cease once the coleoptile tip reaches the soil surface , at which point the coleoptile opens and the seedling emerges. Semidwarf rice cultivars have historically been found to have lower emergence rates from deeper plantings than taller rice cultivars, primarily due to lower rates of mesocotyl elongation . However, semidwarf rice cultivars with higher rates of mesocotyl elongation have been produced in recent years . Though semidwarf California rice cultivars are bred primarily for water-seeding, these cultivars are known to have high germination rates and high vigor, which are essential for emerging through water depths of up to 20 cm . In this way, CA cultivars may be suitable for the increased elongation and emergence needs of drill seeding to depths exceeding 2 cm. The present study aims to evaluate above- and below ground physiological responses of common CA rice cultivars to planting at various soil depths in a controlled environment, and thus identify cultivars that may be suitable for field-level studies on the efficacy of a deep-seeded stale seedbed strategy for herbicide resistance management in California rice.All rice cultivars used in the experiments are nonhybrid semidwarf medium-grain temperate Japonica, supplied from commercially available seedlots with tested germination of 96% or greater. The four cultivars selected are widely planted in California, with full-season durations of about 150 days. ‘M-105’ is a very-early heading cultivar, averaging 83 days to 50% heading, and is grown on approximately 12% of California rice land . M-105’s earlier heading might compensate for delayed emergence under deep seeding conditions. ‘M-205’ and ‘M-209’ are sequential releases of higher yielding cultivars that have the latest average heading times of the tested cultivars, at 94 days and 92 days to 50% heading, respectively. M-205 and M-209 are grown on approximately 11% and 24% of California rice lands, respectively. M-205 and M-209 have higher susceptibility to low-temperature induced sterility than M-105 and M-206. Thus, dry racks for weed deeper planting may provide better temperature moderation and protection from blanking for M- 205 and M-209.
In addition, M-205 and M-209 are larger-seeded varieties than M-105 and M-206; this may provide increased reserves to push through soil when planted deeply. ‘M-206’ is the most widely planted rice cultivar in California, grown on roughly half of the region’s area, with an average time to 50% heading of 86 days. M-206 has the lowest lodging resistance of the tested cultivars; deeper planting should help anchor M-206’s roots better and reduce propensity to lodging.Experiments were conducted over 2017-2018 in glasshouses at the University of California, Davis , Davis, CA., USA, and at the Rice Experiment Station , Biggs, CA., USA. Plastic tubs measuring 43 x 29 x 24 cm were bottom-drilled with 1.5-cm diameter holes to facilitate slow drainage after irrigation events and simulate field drainage in flush-irrigated drill seeded systems. Tubs were pre-filled to set depths with rice field soil from Biggs, Ca., a Yolo clay loam which was carefully leveled. Ten seeds of each cultivar were randomly assigned to be placed in four regularly-spaced longitudinal rows atop the level bed, and subsequently covered with soil to provide a uniform planting depth for each tub. Tubs were irrigated to saturation after seeding, and allowed to drain. Subsequent irrigation flushes to 15 mm water depth occurred every eight days thereafter, and allowed to drain. Supplemental lighting was provided by 1000 w high-pressure sodium lamps or 1000 w metal halide lamps providing 400 µmol m-2 sec-1 photosynthetic photon flux density, with a 16 h photo period. Average day / night temperature and relative humidity were 32 / 18 °C and 44 / 80%, respectively, in both experiments.This study was conducted at UCD. Rice seed were planted as described above to 2.5, 5.1, and 7.6 cm depths. Seeds or seedlings of each cultivar were carefully excavated every two days after planting until 20 DAP, and carefully washed with water. Germination, emergence, coleoptile and mesocotyl length, and total length were recorded. Germination was determined by the presence of an emerged coleoptile of greater than 1 mm length. This experiment was conducted in a split-plot design, with factors of depth as main plot, and cultivar as sub-plot. Treatments were replicated six times, and the study was repeated once.This study was conducted at RES. Rice seed were planted as described above to 0, 1.3, 2.5, 3.8, 5.1, 6.4, and 7.6 cm depths. Seedling emergence was noted on a daily basis, and plant height was recorded weekly, until 28 DAP. At 28 DAP, emerged rice plants were randomly thinned to two plants per cultivar per tub. Plant height and number of tillers were recorded weekly until eight weeks after planting , upon which the number of leaves and above ground fresh weight per plant were recorded. Plants were dried at 50°C for one week, and weighed. This study was conducted in a split-plot design in the same manner as the germination and elongation study. Treatments were replicated four times, and the study was repeated once.No significant run-by-factor effects were detected in either study; therefore, data for each measurement were pooled across runs. Germination data were highly skewed towards 100% and were analyzed by semiparametric one-inflated beta regression using the package “gamlss” in R software . An index of emergence potential was calculated by dividing the proportion of emerged seedlings by the proportion of germinated seedlings, and analyzed as described above. While specific to soil and environmental conditions , this index is a reasonable indicator of vigor differences between cultivars . Cumulative seedling emergence data were analyzed by time to-event analysis using the package “DrcSeedGerm” in R, to account for observation interval censoring, and to account for un-germinated or un-emerged seeds at the time of the study’s termination , and fit to a three-parameter log-logistic model. Total emergence data at 8 WAP were fit with quadratic regression. All other data were analyzed via standard linear regression and ANOVA, using R and JMP® 14Pro , with means separated by Tukey honestly significant difference at α = 0.05, where applicable.Each cultivar achieved 94% or greater germination at all planting depths by 20DAP. The greatest germination was found with cv. M-209, with 97% germination at 2.5 cm planting depth, and the lowest was found with M-105 at 7.6 cm depth, with 95% germination. No significant differences in germination were found between cultivars at any depth , however M-205 and M-209 tended to consistently greater germination at all depths. Germination rates decreased slightly with deeper plantings for all cultivars, however the negative trend was not statistically significant. Emergence potential indices showed marked differences in the proportion of germinated seeds that were able to emerge. M-209 exhibited a greater emergence potential than the other cultivars at each planting depth, and was the only cultivar to have an index greater than 0.95. In addition, the apparent rate of decline in potential with increasing depth was lowest for M- 209.