Research report from OSU's North Willamette Agricultural Research and Extension Center
Delbert Hemphill and Bob Mcreynolds
OSU Dept of Horticulture, NWREC
The major cultural problem in overwintered onion production is weed control. Onions are a slow-growing crop which competes poorly with weeds. Since the crop is in the ground for eight or nine months, and cultivation is nearly impossible during the winter rainy season, both good weed control at planting and good postemergence control are necessary. The weed control task has been made more difficult by the loss of registration of effective preemergence herbicides such as propachlor. The purpose of this trial was to compare the effectiveness of two preemergence herbicides in combination with several alternatives for fall and spring postemergence herbicides. In addition to weed control rating, observations were made on onion injury and stand reduction by the herbicides and onion yield data were obtained.
'Sweet Winter' onions (ARCO Seed Co.) were seeded in 3 rows on 5-foot beds at about 20/foot on August 28, 1985. The seedbed was prepared by rotary tillage following a broadcast application of 10-20-10 fertilizer at 800 pounds/acre, and gypsum at 100 pounds/acre. Either Dacthal at 10 pounds active/acre or Ramrod at 4 pounds active/acre was applied on August 29 and irrigated in. These main plots consisted of 3 beds x 80 feet and were replicated four times in randomized block design.
On October 11 the plots were split by four postemergence ("fall application") treatments. These were oxyfluorfen (Goal) at 0.125 pounds/acre, oxyfluorfen plus cultivation between rows on October 2, pendimethalin (Prowl) at 0.5 pounds/acre, or pendimethalin plus cultivation. The resulting subplots consisted of 3 beds x 20 feet. All plots received an application of oxyfluorfen at 0.125 pounds/acre on January 6, 1986.
The "fall application" subplots were split by "spring" application of oxyfluorfen at 0.125 pounds/acre, chloroxuron (Tenoran) at 3.0 pounds/acre, or ureasulfuric acid (N-Tac) at 15 gallons/acre (30 gallons/acre total volume) on February 21, 1986. The resulting sub/subplots consisted of a 20-foot section of a single bed. On April 8, 1986, all plots received an application of metolachlor (Dual) at 2.25 pounds/acre. All herbicide applications were followed by rainfall or at least 0.5-inch irrigation. An additional 100 pounds N/acre was applied as ammonium nitrate in February and March. Weed control and crop injury ratings were made after each treatment. Bulbs were topped and harvested on July 16, 1986.
Results and Discussion
The preemergence herbicides had no effect on seedling stand (Table 1). Weed control ratings were first made on October 1, 1985. Each bed was scored separately within each main plot. Weed control was clearly far superior with Ramrod as compared to Dacthal (Table 1). Dacthal provided good control of chickweed but little control of Poa annua, groundsel, dog fennel, shepherdspurse, and mustards. Ramrod also provided inadequate control of knotweed and the larger grasses, but these were not a problem in most of the plot area. Ramrod provided very little chickweed control.
Weed control ratings were made again two weeks after the "fall application" of herbicides and cultivation of appropriate plots. For the main effect of the preemergence application, Ramrod was again clearly superior (Table 2). Goal provided superior burn down of escaped weeds, especially shepherdspurse and groundsel, but had no effect on grasses or chickweed. Cultivation had no effect on weed control ratings as the soil moisture was too high to get good weedkill from the cultivation alone. Some plots receiving Ramrod and Goal were essentially weed-free at this point. There were no significant interactions affecting weed control between preemergence and fall herbicide treatments or between the fall-applied herbicides and cultivation. Only main effects are given in Table 2.
Dacthal and Ramrod had no significant effect on onion injury rating, but Goal injured the onions more than did Prowl. The Goal stunted and twisted the onion leaves. Cultivation tended to reduce Goal and Prowl injury slightly, probably by shielding the onion leaves from the herbicide spray or by providing more cover for the roots. There were no significant interactions and only main effects are given in Table 2.
Table 1. Effects of Dacthal and Ramrod at planting on weed control rating on October 1, 1985, and on onion seedling stands Seedlings/foot Weed control ratingZ Dacthal 10.6 2.7 Ramrod 10.9 7.2 NSY * _______________________________________________________________ ZTen point scale, 9 = no weeds, 0 = no control. YNS = no significant difference, ** = difference significant at 1% level. Table 2. Main effects of preemergence and "fall" herbicide applications and cultivation on weed control and onion injury on October 25, 1985 Weed control Onion injuryZ Dacthal 3.0 2.1 Ramrod 5.4 1.8 **Y NS Goal 5.6 2.8 Prowl 2.7 1.1 ** ** Cultivated 4.1 1.8 Non-cultivated 4.2 2.1 NS * ZFive point scale with 0 = no injury, 4 = seedling destroyed. Y**, *, NS: differences significant at 1% and 5% levels, and no significant differences, respectively.
The "spring" herbicide applications were made between February 21 and 26, 1986. There were no interactions between preemergence, fall and spring applications, so means given in Table 3 are main effects only. Weed control was rated on a 0-9 scale with 9 being weed free. Onion stands and vigor were rated on a 0-5 scale, with 5 the most desirable rating. Notes were also taken on weed species in each plot.
Goal spotted grass leaves but did not kill the grasses. Tenoran was ineffective on grasses but controlled chickweed. N-Tac provided no control of established weeds. Ramrod controlled most grasses and Dacthal plots were still free of chickweed. No difference was expected at this time between the fall treatments (Goal vs. Prowl) since all plots had a Goal treatment in January.
Looking at simple effects, the highest mean weed control ratings were obtained with Ramrod + Goal (no cultivation) + Tenoran (8.25), followed by Ramrod + Goal + cultivation + Tenoran (7.6), followed by Ramrod + Prowl + cultivation + Tenoran (7.25), followed by Ramrod + Prowl (no cultivation) + N-Tac (7.0). The lowest mean rating was for Dacthal + Goal (no cultivation) + N-Tac (1.5), followed by Dacthal + Prowl + cultivation + Tenoran (2.5). The best combinations using Dacthal at planting were Dacthal + Goal (no cultivation) + Tenoran (5.5), and two with a score of 5.0: Dacthal + Goal + cultivation + Goal and Dacthal + Prowl + cultivation + Tenoran.
Ramrod and Goal both reduced the stand rating and reduced onion vigor. Lowest vigor was on plots receiving Ramrod and two or three applications of Goal. There was a strong negative correlation between weed control rating and onion vigor rating (Rxy = -0.638, p=0.001). However, the very best vigor ratings did not coincide with the worst weed control. Vigor was usually highest on plots with high populations of low growing Poa annua or chickweed, but few large weeds.
Table 3. Weed control, stand, and vigor ratings on March 17, 1986 Herbicide Weed Onions/ Onion Major weeds control foot vigor present Dacthal 3.8 3.1 3.0 Poa, grasses, groundsel Ramrod 6.6 2.6 2.4 chickweed, groundsel, Poa ** ** ** Goal 5.2 2.0 2.1 Poa,grasses, chickweed Goal+Cultivation 5.6 2.3 2.5 Poa, grasses, chickweed Prowl 4.7 3.4 2.9 Poa, grasses, groundsel, chickweed Prowl+Cultivation 5.3 3.7 3.3 Poa, grasses, groundsel, chickweed NS ** ** Goal 5.2 2.6 2.4 Poa, grasses, chickweed Tenoran 5.6 2.9 2.7 Poa, grasses, groundsel N-Tac 4.9 3.0 3.0 Poa, chickweed, grasses, groundsel NS NS *
The reduced stand ratings with Ramrod at planting (Table 3) or Goal applied in the fall were reflected in lower numbers of bulbs present at harvest (Table 4).
The main effect of Ramrod at planting, when averaged over the other herbicide treatments, was to slightly reduce the total number of bulbs, increase the number of large bulbs, and increase mean bulb weight, grade 1 yield and total yield as compared to Dacthal (Table 4). Since the stand reduction with Ramrod was very small, the yield increase can be attributed to improved weed control.
Goal, when compared to Prowl, reduced stands by more than 50 percent, and decreased both total and grade 1 yield. Mean bulb weights were greater with Goal. This increased bulb weight could be caused either by reduced competition among onion plants or by superior weed control.
Cultivation in the fall prevented stand reduction and tended to increase bulb weight, resulting in highly significant increases in total and grade 1 yield. Cultivation may have shielded the onion plants from spray damage. Prowl produced a higher number of grade 1 bulbs/plot than did Goal when following Ramrod, but not when following Dacthal (Table 5). This can be attributed to the relatively greater impact on weed control of Goal following Dacthal than Goal following Ramrod.
Main effects of the spring-applied herbicides on yield were not significant (Table 4), but the trend was toward greatest yields with Tenoran. There were interactions of herbicides at planting with spring-applied herbicide affecting several components of yield (Table 6). Total yield and number of grade 1 bulbs/plot were highest with Tenoran following Ramrod but not following Dacthal. Since the major effect of Tenoran was to burn down chickweed and Ramrod-treated plots contained more chickweed than did Dacthal-treated plots, the greater effect of Tenoran following Ramrod was to be expected.
The highest yielding treatments were Ramrod + Prowl + cultivation + Tenoran, Ramrod + Prowl + cultivation + Goal, Ramrod + Prowl + Tenoran, and Dacthal + Prowl + cultivation + Goal (Table 7). Each of these except the last produced much better than, average weed control. The correlation of total yield and weed control rating made in March was not strong, however (Rxy =0.291, p=0.085). This is to be expected since stands were often reduced with those treatments producing excellent weed control. The yield of grade 1 bulbs (Rxy =0.35, p=0.005) and mean bulb weight (Rxy =0.63, p=0.001) both correfXted strongly with weed control rating. xYhe major weeds present at harvest were grasses, dog fennel, and chickweed (except on Tenoran-treated plots).
In summary, these results confirm that loss of use of Ramrod and Tenoran as onion herbicides greatly increases the difficulty of growing a successful crop of overwintered onions. Goal is promising as a postemergence burn down treatment but may injure the crop. N-Tac provided little weed control in this trial, but its contribution to the N needs of the crop must be considered. The Prowl used in this trial was old, with crystals precipitating out. This material deserves further investigation.
Table 4. Main effects of weed control programs on yield and bulb size of overwintered onions, July 16, 1986 Yield Mean bulb wt. No. of bulbs/plot Timing Herbicide Grade 1 bulbsZ All bulbs Grade 1 All Grade 1 All __________________________________________________________________________________ --------tons/acre-------- ----ounces--- Planting: Dachthal 1.5 9.6 19.5 2.9 3.4 89.4 Ramrod 4.5 14.9 10.0 4.7 11.2 81.2 **Y ** NS ** ** NS Fall: Goal 2.4 8.4 10.6 4.1 5.7 55.5 Prowl 3.5 16.1 9.6 3.5 8.9 115.1 * ** * * * ** Cultivated 3.9 15.0 10.2 3.9 9.7 96.8 Non-cultivated 2.0 9.6 10.0 3.6 4.9 73.9 ** ** NS NS ** * Spring: Goal 3.0 12.1 10.4 3.9 7.2 82.5 Tenoran 3.6 13.6 9.9 3.8 8.8 89.2 N-Tac 2.4 11.7 9.9 3.6 5.8 88.2 NS NS NS NS NS NS ZOver 3-inch diameter. Y**, *, NS: Means differ significantly at 1% and 5% levels, and no significant differences, respectively. Table 5. Interaction of herbicides at planting and fall herbicide applications on number of grade 1 bulbs/plot, July 16, 1986 Herbicide No. of grade Planting Fall 1 bulbs/plot Dacthal Goal 2.0 Goal + cultivation 5.1 Prowl 2.6 Prowl + cultivation 4.0 Ramrod Goal 4.6 Goal + cultivation 11.0 Prowl 10.4 Prowl + cultivation 18.8 LSD (0.05) 5.6 Table 6. Interaction of herbicides at planting and spring herbicide applications on total onion yield, mean weight of grade 1 bulbs, and number of grade 1 bulbs/plot, July 16, 1986 Herbicide Total yield Mean wt. of No. of grade 1 Planting Spring (tons/acre) grade 1 bulbs (ounces) bulbs/plot Dacthal Goal 11.0 10.6 4.8 Tenoran 8.1 10.4 2.5 N-Tac 10.4 9.3 3.0 Ramrod Goal 13.1 10.1 9.6 Tenoran 19.1 9.4 16.0 N-Tac 13.0 10.5 8.4 LSD (0.05) 4.0 1.0 4.9 Table 7. Simple effects of all treatment combinations on yield of grade 1 bulbs, total yield, and mean bulb weight, July 16, 1986 Herbicide Total yield Grade #1 yield Mean bulb Planting Fall Spring wt. ----------tons/acre--------- ounces Dacthal Goal Goal 4.7 1.3 2.6 Tenoran 1.5 0.3 2.3 N-Tac 4.7 0.5 2.0 Goal+Cult. Goal 2.8 1.4 6.5 Tenoran 6.7 1.2 2.2 N-Tac 9.1 2.7 3.1 Prowl Goal 12.6 1.6 2.6 Tenoran 6.7 0.7 3.4 N-Tac 11.7 0.3 2.0 Prowl+Cult. Goal 16.6 2.6 3.2 Tenoran 12.0 1.2 2.3 N-Tac 9.0 0.5 2.6 Ramrod Goal Goal 6.6 2.9 5.1 Tenoran 4.3 2.8 5.6 N-Tac 6.7 3.1 4.0 Goal+Cult. Goal 9.0 4.7 4.4 Tenoran 15.3 5.8 5.8 N-Tac 14.5 3.8 4.8 Prowl Goal 9.9 1.5 3.3 Tenoran 17.2 5.0 3.9 N-Tac 13.9 4.1 4.5 Prowl+Cult. Goal 24.5 8.3 4.6 Tenoran 25.8 6.6 4.6 N-Tac 9.4 2.1 3.8 LSD (0.05) 8.0 4.0 2.7