Floodwaters are generally maintained at 10-20 cm for the entire season. California rice cultivars are temperate japonica inbred lines, with sufficient vigor to quickly elongate and escape deep floodwaters . Water seeding was adopted in the region in the 1920’s to suppress competitive grass weeds , and has remained the preferred method of rice cultivation in California, even as herbicides have been widely available for decades . Although the WS rice cropping system is optimized for the region, it is not without disadvantages. Water seeding encourages higher seeding rates that can incur higher production costs, because floating seedlings are subject to wind drift and predation, both of which may result in reduced or patchy stands at lower seeding rates. Surface-rooted rice is also prone to lodging. Irrigation water usage is also of concern , as California is regularly beset by drought and irregular rainy seasons exacerbated by climate change. The near-exclusive use of permanently-flooded rice culture has also resulted in a small spectrum of well-adapted and competitive grass species , as well as aquatic broad leaves and sedges . As the permanently-flooded cropping system essentially precludes cultural weed management practices such as cultivation, and as large farm size and high labor costs discourage hand-weeding, herbicides are the sole means of weed control outside of water management for most California growers . Although effective herbicides have been available for California rice since the 1960’s, the nearly exclusive use of water seeding has meant that the number of registered active ingredients remains low, amid pesticide contamination concerns and California’s stringent regulatory structure . This limited herbicide palette restricts herbicide rotation. Since rice is largely grown year after year in the region, vertical grow room design the combined effects of a water-seeded monoculture, and extensive use of limited available herbicides on a small weed spectrum, have resulted in widespread cases of herbicide resistance .

Cultural methods for weed and resistance management in California rice are generally limited to modifications of the dominant WS system . For example, with the “stale seedbed” method rice fields are prepared for planting as usual, but are flushed with water prior to seeding to promote weed germination . Nonselective herbicides are typically used as a burndown treatment on emerged weeds , and fields are then flooded and air-seeded as usual. This method can be a useful strategy to manage weeds that are resistant to registered rice herbicides, by introducing novel nonselective herbicides without known local resistance . However, implementing a stale seedbed can delay rice planting and shorten the growing season, potentially depressing yields . Stale-seedbed can be followed by drill seeding to shorten the delay between burndown treatment and rice planting. In DS systems, rice is dry-drilled to 1.25 – 2 cm, and fields are flush irrigated intermittently as the stand develops and herbicides are applied, then flooded for the remainder of the season 30-40 days after seeding . This method discourages aquatic weeds and algae , however in this system the rice often emerges synchronously with grass weeds , reducing the stand’s ability to compete with weeds. This also limits management of resistant weeds to the short preplant burndown window. However, if rice seed is sown to depths exceeding 2 cm, it might emerge later than early-germinating grasses , thus allowing cultural or chemical weed management practices to be used safely on emerged weeds without injuring the rice . This would lengthen the stale-seedbed burndown window, allowing more weeds to be managed by the burndown treatment. As California rice cultivars are bred for water seeding, they have suitable vigor to emerge through water depths of up to 20 cm . This high vigor may make California rice cultivars suitable for drill seeding to depths greater than 2 cm.

If planting vigorous rice deeply can permit delayed, but even rice stand emergence, it should be possible to combine drill seeding with a stale seedbed as an integrated approach to herbicide resistance management. This “stale-drill” method could permit the use of a novel mode of action in a post plant-burndown treatment, which would safely manage key herbicide resistant weeds prior to stand emergence, without crop injury or delayed planting. Studies under controlled conditions comparing several California rice cultivars’ responses to burial depth indicated varying levels of vigor between cultivars . The cultivar ‘M-209’ was found to have the greatest vigor of those tested, in terms of below-soil elongation, emergence, and early season development. The purpose of this study was to compare stand establishment and yield components of M-209 and the most commonly planted cultivar, ‘M-206’, when seeded at two different depths. In addition, herbicide programs featuring a PPB application were evaluated for optimized late-season weed control.Studies were conducted in a split-split-plot design, with planting depth as the main plots, cultivar as the subplots, and herbicide treatment as the sub-subplots, with three replicates each year . Main plots were 16 x 18 meters, and were surrounded by 2.2-meter wide levees to allow independent flush-irrigation and flooding. Planting depths in main plots were either 3 or 6 cm. Within the main plots, cultivars ‘M-206’ and ‘M-209’ were planted on approximately half of the plot area each, separated by a 1.5 m unplanted buffer strip. M-206 is the most commonly planted cultivar in California , grown on roughly half of the planted area, with a heading time of 86 days . M-209 is a newer, higher-vigor cultivar that with a heading time of about 92 days . Rice was dry-drilled at a rate of 121 kg ha-1, using a mechanical seed drill with 17.8 cm row spacing. Planting dates were 28 May 2018 and 19 June 2019 .Flush-irrigation for main plots was done by powered pumps. Main plots were flushed immediately after planting, and water was allowed to infiltrate the dry soil.

Subsequent flushes were applied to each main plot independently, as the soil dried and cracks appeared. After final herbicides treatments were applied, the entire field was flooded to 10 cm average water depth for the remainder of the season. Harvest dates for 2018 and 2019 were 20 October and 29 October, respectively.Herbicide programs were evaluated for PPB efficacy, and differences of control for later-emerging weeds. Sub-subplots of herbicide treatments were applied in 3 m x 6 m zones . Three herbicide treatments plus an untreated control were used . Herbicides were applied with a 6 m boom sprayer with six 8003XR flat-fan nozzles , CO2-pressurized and calibrated for 187 L ha-1 carrier volume. At the date of first observed rice emergence, treated plots received a post plant burndown application of glyphosate at 870 g a.e. ha-1 + 2 % w/v ammonium sulfate . Follow-up early-postemergence and mid-post emergence treatments were applied at 3- leaf and 4-leaf rice stages, respectively. EPOST treatments consisted of bispyribac at 37 g a.i. ha-1 + 0.4% v/v organosilicone surfactant , applied alone or with a tankmix partner of pendimethalin at 1070 g a.i. ha-1 or clomazone at550 g a.i. ha-1. MPOST treatments were cyhalofop + 2.5% v/v crop oil concentrate.The study site weed seedbank was previously described in Brim-DeForest et al. and BrimDeForest et al. . Weed control evaluations measured the early-season efficacy of PPB treatments in this program, clone rack as well as the contributions of PPB treatments to overall control. The potential for later applications of pre-emergent herbicides to enhance control of lateremerging weeds was also investigated. Weed responses to herbicides and treatment timing differences imposed by rice planting depth were measured. Weed density in each plot after PPB treatment was estimated 20 days after planting by counting plants in 30 cm x 30 cm quadrat samples, with three averaged subsamples per plot. Follow-up weed density counts were performed at 45 and 70 DAP, following the same methodology. Echinochloa spp. and sedges were grouped in their respective genera for quadrat counts.Rice growth and development in response to herbicide program and planting depth were measured throughout the season. Of particular interest were crop responses to PPB applications of glyphosate, as well as the effects of planting depth and weediness on crop development and yield components. Date of rice emergence was determined by visual estimation, and defined as >10%of rice plants visible at the soil surface, and was used to time PPB treatment.

Rice stand density was recorded at 21 days after planting by counting plants in 30 cm x 30 cm quadrats, with three subsamples averaged per plot. Due to different maturation rates of the cultivars used in this study, tiller density was recorded at 90 DAP and 110 DAP by counting tillers in 30 cm x 30 cm quadrats, with three samples per plot. Time to 50% heading was estimated visually, and plant heights were recorded with a meter-stick at 120 DAP. Prior to field harvest, ten panicles per plot were randomly selected, hand-harvested, and dried for three days at 50°C. Grain yield per panicle and 1000-grain weight were measured, and adjusted to 14% moisture content. Filled and total florets per panicle were measured, and percentage of unfilled florets was calculated. Whole plots were harvested and yields measured with a small-plot combine with a swath width of 2.3 m, and were adjusted to 14% moisture content.All data –with the exception of time-to-heading– were subjected to ANOVA and linear regression analyses using the agricolae and emmeans packages in R , and JMP® 14Pro , using planting depth, cultivar, and herbicide treatment as fixed effects, and replicates as random effects. Significant year-by-depth and year-by-treatment and interactions for all data were observed; therefore, data were re-analyzed separately by year, using the same fixed and random effects as described above. In both years, no differences were observed between rice planting depths, cultivars planted, or among applied herbicide treatments for weed count data, therefore data for planting depth, cultivar, and treated plots were pooled and re-analyzed as treatment versus UTC . Similarly, in both years, no differences were observed between herbicide treatments for rice stand and yield components, therefore treated-plot data were pooled and re-analyzed as treatment versus UTC . Pooled data met assumptions of homogeneity and normality of variance, and were untransformed for analysis.Rice became visible at the soil surface by 7 DAP in 2018, and 6 DAP in 2019, by which time grass and sedge seedlings were very dense in all plots. At date of emergence either year, M-209 had slightly greater emergence than M-206 at both planting depths. Likewise, rice planted to 3 cm was slightly taller than rice planted to 6 cm, regardless of cultivar. Nevertheless, at DOE there were no differences in emerged seedling heights between cultivars or planting depths either year, and stands in all plots were even. In both years, glyphosate PPB was applied at DOE. Rice firstleaf tips exposed to the PPB treatment died off within a few days, however plants developed normally and exhibited no stunting, chlorosis, or other injury symptoms . Comparing pooled herbicide treatments with UTC plots, rice stand reductions were observed by 21 DAP in 2018 . Averaged across planting depths and cultivars, rice plants m-2 were reduced by 46% in UTC plots in 2018 , and were eventually reduced to zero. In 2019, no stand reductions were observed in UTC at 21 DAP, however significant reductions in all other growth and yield components were observed in UTC plots. Rice stand development in treated plots was affected by cultivar and planting depth both years. In 2018 stand density for M-206 and M-209 planted at a 6 cm depth was reduced by 15.4% and 5.2% , respectively, relative to the 3 cm planting depth. However, stand density was not affected by cultivar or planting depth in 2019. Increased tillering in M-206 compensated for stand reductions in the 6 cm planting depth in 2018; M-206 tillering decreased only 3.2 % between the 3 cm and 6 cm planting depths. For either cultivar, tiller density decreased slightly at 6 cm planting depth in 2018, and increased slightly in 2019. Cultivar differences in tillers plant-1 were only observed in 2018, averaging 2.3 and 1.9 tillers for M-206 and M-209, respectively. Time to heading was affected by cultivar in 2018, and by both cultivar and planting depth in 2019 . Time to 50% heading in 2018 was 75 DAP and 83 DAP for M-206 and M-209, respectively. In 2019, T50 for M-206 was 76 DAP for both plating depths.