Overwintered onions in the Willamette Valley are seeded in early September and harvested in the following spring or summer. Obtaining strong and early growth in the spring is essential to achieve large bulb size and profitable yields. However, soil and air temperatures are usually less than optimal during the spring growth period, possibly limiting response to fertilizers.
Highest yields have been obtained with a total N application of 200 to 300 pounds/acre, with the majority of the N applied in the spring. No information is available on potential differences in overwintered onion response to type of spring-applied N; e.g., NO3- vs. NH4-N. On cold soils with limited conversion of other forms of N to plant available NO3, applying a NO3-N fertilizer might improve N use efficiency. However, NO3 is also not tightly bound to soil particles and may be quickly leached through the root zone by heavy spring rains.
Since P availability is limited on cold soils, overwinter onions might also respond to banded application of P fertilizers or to lime, which increases P availability. Onions also have a high S requirement, but overwinter onion response to a fertilizer S source has not been studied in the Willamette Valley.
The following experiments were designed to evaluate the effects of lime, spring-applied P, spring-applied CaSO4 (gypsum), and two N sources on yield and elemental concentrations of overwintered onions.
Agricultural limestone (95% CaCO3 equivalent) at 0, 2, 4, and 6 tons/acre was applied in 1979 to 2,300 square feet plots of Willamette silt loam with four replications of each treatment in randomized block design. Resulting soil pH at planting in 1982 averaged 5.5, 6.0, 6.2, and 6.6, respectively. Raised beds (8 inches high, 5.5 feet wide) were formed in early September, 1982 following a broadcast application of 800 pounds/acre of 10-20-10 and seeded with 3 rows/bed of OWY 100 (ARCO Seed Co.) onion on September 15.
Propachlor herbicide was applied at 4 pounds/acre on September 16, and again on October 20 and December 10. Many weed species escaped the propachlor treatment. Linuron at 1.0 pound/acre was applied on February 2, 1983 and chloroxuron at 3.0 pounds/acre was applied on March 21. These treatments eliminated most established weeds; large grasses and plantain were pulled by hand in late spring. No further herbicides were applied and plots were hand-hoed in May, primarily to control vetch and groundsel.
On Febrary 15, 1983, the lime main plots were randomly split into six subplots (one bed x 24 feet) by application of the following: 1) ammonium nitrate at 50 pounds N/acre, 2) ammonium nitrate as above plus 100 pounds gypsum/acre, 3) ammonium nitrate as above plus 100 pounds 0-45-0/acre, 4) ammonium nitrate, gypsum, and 0-45-0, as above, 5) ammonium sulfate at 50 pounds N/acre, 6) check (no spring fertilizer). The N materials were reapplied at 50 pounds N/acre on March 28 and May 13, 1983; concentrated superphosphate (0-45-0) and gypsum were not reapplied. Leaf samples were collected for plant tissue analysis on April 26 from plots representing three replications of the zero and 4 tons/acre lime rates and all subplot treatments. On April 27, stand counts were made for all treatment combinations. Onions were topped and harvested from the center row of each plot on July 5, 1983. Bulbs were graded into #1 (more than 3 inch diameter) and #1 categories before weighing.
Methods were as above for 1983, except as follows. Seeding date was September 13, 1983; variety was ARCO Sweet Winter. Propachlor was applied four times and only chloroxuron was applied in the spring. Metalaxyl was applied twice for mildew control. Subplot treatments were applied on February 17, 1984, when half of each main plot was sidedressed with 0-45-0 at 150 pounds/acre. Ammonium nitrate was applied to all plots at 50 pounds N/acre on February 17, March 26, and May 3. Harvest was on July 24, 1984.
Results and Discussion
Application of lime significantly increased onion stands, total onion yield, yield of #1 bulbs, mean bulb weight, and number of bulbs harvested (Table 1). Stand counts were made in April 1983, too late to determine whether the lime application enhanced onion germination and emergence or enhanced survival by increasing the growth rate of seedlings. Previous experiments on Willamette soil indicated that liming increases seedling emergence of onions and several other small seeded vegetable crops. Since the application of lime also visibly stimulated early plant growth (no measurements recorded), enhanced winter survival of larger seedlings also may have contributed to the effect of lime on onion stands. Of the subplot treatments, only N application affected stands, with a small, but statistically significant, increase in stand on plots which received no spring fertilizer. Since no stand counts were made before application of the first subplot treatments, it cannot be determined whether spring fertilizer application actually caused some stand reduction or the stand differences reflected existing variability within main plots.
Most of the stand and yield response to lime occurred with application of only 2 tons/acre; however, further significant increases in yield and number of grade #1 bulbs were obtained at 4 tons/acre (Table 1). Leaf tissue of plants grown on limed soil contained significantly higher concentrations of P, K, and Ca, and significantly lower concentrations of Zn and Mn than did leaf tissue grown on unlimed soil (Table 2). Since P levels were quite low compared to reported values and were increased 24% with application of 4 tons/acre of lime, much of the yield response to lime might be ascribed to increased P availability. However, increased K and Ca uptake or reduction of Mn toxicity may also have been involved in the lime response.
Application of spring N fertilizer, when averaged across lime, P, gypsum, and form of N applied, significantly increased total and #1 yields and bulb weight (Table 1). There were no significant N x lime interactions, and highest overall yields (30-33 tons/acre) were obtained with combinations of the highest rate of lime and spring application of either ammonium nitrate or ammonium sulfate. Application of spring N increased leaf tissue N, Zn, and Mn concentrations. Undoubtedly, the yield response to N was primarily attributable to increased soil N supply. The increase in tissue Zn concentration may also have played a role since Zn concentration of plants which received no spring fertilizer application was low compared to values reported in the literature. The increase in leaf tissue Zn and Mn concentrations with spring N application may have been caused by a temporary localized decrease in soil pH after application of the acidifying N fertilizers.
Within the subset of plots receiving a spring application of fertilizer, there was a trend toward higher yields and mean bulb weights with ammonium sulfate as N source (Table 1). These differences were never significant at the 95% level; however, the increase in total yield was significant at the 90% level. Leaf tissue N levels were slightly higher with ammonium sulfate as N source, but other tissue elemental concentrations were not significantly affected by N source (Table 2). Certainly, it does not appear necessary to provide NO3-N to assure good onion yields.
Within the subset of treatments receiving a spring application of ammonium nitrate, application of concentrated superphosphate did not affect overall yields but did slightly increase mean bulb weight (Table 1). There were no significant P x lime interactions. Application of P had no effect on leaf elemental concentrations (Table 2). Since P had no effect on tissue P levels, it is evident that the surface application did not bring the relatively insoluble P into sufficient contact with the root mass, or that some other factor prevented effective uptake. Lack of P uptake from the fertilizer probably precluded any yield response. However, since the winter and spring were unusually mild, any P effect on yield may have been masked by better than normal spring growth on all plots.
Also within the subset of plots fertilized with ammonium nitrate, application of gypsum increased total and #1 yields. Some of this increase was caused by slightly higher (increase not statistically significant) stands on subplots fertilized with gypsum. Mean bulb weight, however, also increased with gypsum application and was the major component of the yield increase (Table 1). Leaf tissue concentrations of the analyzed elements were not affected by gypsum application. Leaf S levels were not measured, but it is possible that the response to gypsum was caused by increased S availability. Sulfur availability may also have been involved in the nearly significant yield increase with ammonium sulfate compared to ammonium nitrate.
There were no significant P x gypsum interactions affecting yield or leaf elemental concentrations, but lime and gypsum interacted strongly in increasing yield of #1 onions (Table 3). The greater response to gypsum at higher rates than at lower rates of lime, and the tendency for yields to increase with the last increment of lime in the presence but decrease in the absence of gypsum, indicate that S uptake may be a limiting factor in onion production at near neutral soil pH. There were no significant lime x gypsum interactions affecting leaf elemental concentrations.
Table 1. Main effects of lime, spring-applied N, form of N, spring-applied P, and gypsum on yield parameters of overwinter onion, 1983 Stand Total Yield of #1 bulbs Mean bulb wt. Mean bulb onions/ yield #1 onions harvested/ all onions wt., #1's plot (T/A) (T/A) plot (oz) (oz) Lime (T/A) 0 92 7.6 3.1 17 2.7 8.0 2 127 22.3 13.7 66 5.6 8.9 4 129 25.1 16.7 82 6.0 8.9 6 129 26.5 18.6 89 6.2 9.1 LSD(0.05) 19 3.1 3.6 9 0.8 0.5 Am. nitrate 117 21.0 14.1 45 5.4 8.9 Am. sulfate 125 23.3 15.4 49 5.6 8.9 NSZ NS NS NS NS NS +P 116 20.8 14.2 45 5.4 9.2 -P 118 21.2 13.9 46 5.2 8.7 NS NS NS NS * * Gypsum 123 22.3 15.4 48 5.6 9.0 Gypsum 112 19.8 12.8 42 5.1 8.8 NS * ** * * * +N 119 21.5 14.4 46 5.4 8.9 -N 125 14.9 6.5 24 3.8 8.0 ** ** ** * ** ** Z**,*,NS: significant at 1% and 5% levels, and non-significant respectively. Table 2. Main effects of lime, spring-applied N, form of N, spring-applied P, and gypsum on onion leaf tissue elemental concentrations, 1983 Treatment N P K Ca Mg Zn Mn Cu --------------%---------------- -----ppm------ Lime, 0 T/A 2.87 0.136 2.00 0.59 0.148 15.4 93 3.9 4 T/A 2.87 0.168 2.38 0.69 0.157 14.0 52 3.7 NSZ * * * NS * * NS +N 2.97 0.150 2.22 0.64 0.153 15.3 77 3.8 -N 2.37 0.160 2.05 0.63 0.148 11.8 48 3.5 ** NS NS NS NS ** ** NS Ammonium nitrate 2.94 0.152 2.20 0.64 0.153 15.3 74 3.8 Ammonium sulfate 3.09 0.143 2.30 0.64 0.157 15.2 91 3.8 * NS NS NS NS NS NS NS Gypsum 2.97 0.151 2.24 0.65 0.153 15.3 76 3.8 Gypsum 2.91 0.153 2.16 0.64 0.153 15.2 71 3.8 NS NS NS NS NS NS NS NS +P 2.94 0.153 2.22 0.64 0.152 15.4 78 3.8 -P 2.94 0.151 2.18 0.64 0.154 15.2 70 3.8 NS NS NS NS NS NS NS NS Z**,*,NS: significant at 1% and 5% levels, and non-significant, respectively. Table 3. Interaction of lime and gypsum on yield of overwinter onion, 1983 Lime rate (T/A) Gypsum (lb/A) Total yield (T/A) Grade #1 yield (T/A) 0 0 6.4 2.9 100 6.9 2.4 2 0 22.7 14.3 100 24.8 15.7 4 0 25.3 17.5 100 27.5 20.1 6 0 24.5 16.5 100 30.1 23.1 LSD(0.05) 5.2 4.6
Onion yields obtained from all treatments were very low compared to those recorded in previous years. For example, the highest yielding lime treatment produced only 3.6 tons/acre (Table 4) compared to 26.5 tons/acre in 1983. This was due primarily to losses in plant stand which occurred during the severe freeze of December 23-24, 1983. Temperatures as low as 6°F were accompanied by 20 to 40 mph winds, resulting in plant breakage and dessication.
Nevertheless, the surviving plants responded to treatment much as they did in 1982-83. Highest total yields and by far the highest weight of #1 bulbs were obtained with the 6 tons/acre lime treatment. More bulbs were harvested from limed plots, perhaps because of better plant survival or better initial stands. Mean bulb weight also increased markedly with increasing rate of lime. Sidedressing concentrated superphosphate in early spring tended to increase total and #1 yield and mean bulb weight, but these differences were not statistically significant. Mean weight of #1 bulbs did increase significantly with P application, as in 1983. Also as in 1983, there were no significant P x lime interactions.
Because of reduced stands, the gypsum and N source portions of the 1983 experiment were not repeated. Confirmation of these effects awaits the 1984-1985 trials.
Table 4. Mean effects of lime and spring-applied P on yield of overwintered onions, 1984 # bulbs Total Yield of #1 Mean bulb wt. Mean bulb wt. Treatment harvested/plot yield (T/A) bulbs (T/A) all onions (oz) #1`s (oz) Lime (T/A) 0 22 0.4 0.0 1.2 - 2 55 1.6 0.4 2.6 6.8 4 41 1.5 0.6 3.1 6.4 6 56 3.6 2.3 5.6 8.4 LSD(O.05) 15 2.3 0.9 1.7 1.8 +P 43 1.8 0.9 3.4 7.6 -P 44 1.6 0.8 2.9 6.9 NSZ NS NS NS ** ________________________________________________________________________________________ Z**, NS: significant at 1% level and non-significant, respectively.