Introduction
A soft rot of the broccoli head, caused by the bacterium Erwinia carotovora subsp. carotovora is one of the most devastating diseases of broccoli. The occurrence of this disease is influenced by the presence of the bacterium, free moisture on the florets, and temperatures ideal for bacterial growth. Past research at the Center has studied the effects of plant populations and irrigation practices on the incidence of the disease. Although much valuable information has been gained that may aid growers in avoiding the disease, no cure has been forthcoming. Recent research in Virginia indicated that nitrogen source might play a role in disease incidence. This might be due to differences in the companion ion accompanying the nitrogen, as various N sources also provide S, Ca, K, Na, or P to the crop. High tissue levels of Ca or the monovalent cations may help prevent the disease.
Broccoli growers in the Willamette Valley use high rates of N fertilizers, often exceeding 250 to 300 pounds of actual N per acre per season. While the common experience has been that these rates are necessary to achieve maximum yields and quality, our past research has indicated increased incidence of head rot at high rates of N. In addition, a considerable portion of this applied fertilizer is not actually taken up by the crop. This has raised concerns that the remaining nitrogen may be contributing to nitrate pollution of groundwater. Improved efficiency of nitrogen management in broccoli may be possible if the fertilizer could be applied at time of maximum crop need. The purpose of these trials was to study the response of broccoli yield and head rot incidence to a wide range of rates and sources of nitrogen fertilizer. Additional objectives were to evaluate the effect of applied calcium chloride on broccoli yield and quality and to evaluate the efficacy of copper compounds and selected bacterial antagonists for control of broccoli head rot.
Methods
In the first experiment of 1989, 'Gem' broccoli was seeded on 31 March in 31 cm3 plugs (Landmark Plastic Corp. 128-cell trays) in an unheated greenhouse and transplanted on 1 May to a Willamette silt loam, pH 5.4, to which had been applied 500 pounds per acre of a 10N-8.7P-16.7K fertilizer, 0.75 pounds trifluralin, 1.3 pounds chlorpyrifos and 2.0 pounds boron per acre. Carbaryl and diazinon were applied for looper control at 1.0 pound per acre, each, on 25 May. Treatments consisted of a factorial combination of six N sources (ammonium nitrate, calcium nitrate, urea, sodium nitrate, potassium nitrate, and sodium-potassium nitrate); two rates of total N (150 and 200 pounds per acre); and three rates of foliar-applied calcium chloride (0, 10, and 20 pounds Ca per acre). Each plot consisted of a four-row bed, 15 feet long, with plants on 16 x 9-inch spacing. Treatments were replicated four times. The N sources were sidedressed on 29 May and again on 13 June, with half the total rate applied on each date. The calcium chloride was applied on 13 June. An aqueous suspension of Erwinia carotovora was sprayed onto the plants on 17 June and again on 29 June. The plots were harvested on 29 June and again on 7 July.
In the second experiment of 1989, 'Gem' was direct-seeded into Willamette silt loam on 21 June in 16-inch rows and thinned to an average within-row spacing of 9 inches. Treatments consisted of a factorial combination of the above N sources, with the addition of ammonium sulfate, and the same three rates of calcium chloride. Plot size was a four-row bed, 20 feet in length. Treatments were replicated four times. No preplant fertilizer was used; pesticide treatments were similar to those in the first experiment. The total N applied was split with 50 pounds N per acre applied on 7 July, and 75 pounds, each, on 26 July and 21 August. The calcium chloride treatments were applied on 21 August. Boron was applied at 2.0 pounds per acre on 27 July. Erwinia was applied on 21 August and again on 5 September. Plots were harvested on 5 and 12 September. Leaf samples were taken for plant tissue analysis on 5 September and samples of the surface centimeter of soil were taken for pH determination on 19 September.
In a third experiment in 1990, six potential head rot-control treatments were applied to 20-foot sections of two-row beds in a commercial field on 29 September. The same six treatments were reapplied on 6 October and were also applied in a second field on the same date. The treatments, replicated six times in each field, included three possible bacterial antagonists and Kocide at 4.0 pounds per acre. Each of these treatments also included inoculation with Erwinia at 106 cells. The fifth and sixth treatments consisted of Erwinia alone, and an untreated control, respectively.
The applications on 29 Sept. were made with 3XD4 hollow cone nozzles at 1.0 foot spacing and 40 psi, and 250 ml spray solution per plot. The applications on 6 October were made with 8003 T-jet nozzles at 28 psi and 333 ml spray solution per plot. Head rot ratings were made on 11, 16, 20, and 27 October.
In the first experiment of 1990, greenhouse-grown 'Gem' was seeded on 27 March and transplanted on 1 May. Plot preparation included a broadcast and incorporated application of triple superphosphate at 200 pounds per acre, trifluralin at 0.75 pounds per acre, and chlorpyrifos at 1.3 pounds per acre. Propachlor (4 pounds per acre) and boron (2 pounds per acre) were applied immediately after transplanting.
The N sources were ammonium nitrate, calcium nitrate, potassium nitrate, sodium nitrate, and urea. These were applied in factorial combination with three N rates (100, 175, and 250 pounds per acre) and two rates of calcium chloride (0 and 30 pounds Ca per acre) in randomized complete block design with four replications. The initial N application of 50 pounds per acre was made two days after planting. The remaining N was applied three weeks later. Plot size was 16 feet with four rows on an 80-inch bed. Within-row spacing was nine inches. Plots were sprinkler-irrigated as necessary. Harvests were 3 and 10 July.
In the second experiment of 1990, 'Gem' was direct-seeded on 18 July with target plant population equal to that in the first planting. Cultural practices were the same as for the first planting. Treatments consisted of a factorial combination of two N sources, sodium nitrate and urea, with four rates of total applied N (100, 175, 250, and 325 pounds/acre). The first 50 pounds N per acre was applied at planting, the remainder four weeks later. In addition to the above factorial combination, an additional set of plots received either 250 pounds urea per acre at planting or 50 pounds urea per acre at planting and 200 pounds per acre four weeks later. Plots were harvested on 18 and 25, October, and 1 November.
In the final experiment of 1990, 'Gem' was direct-seeded on 25 July, again with the same target populations and cultural practices. Treatments consisted of a factorial combination of three N sources (sodium nitrate, potassium nitrate, and urea) with three rates of applied N (100, 175, and 250 pounds per acre). The first N application was on 1 August, with the remainder applied four weeks later. Plots were harvested on 30 October and 6 November.
Results
In the first experiment of 1989, neither head weight, head width, percent good heads, percent head rot, nor percent downy mildew-affected heads was significantly affected by N source (Table 1). However, the incidence of head rot tended to be lower on plots receiving either potassium or sodium nitrate. This trend, though not statistically significant, was particularly strong at the first harvest. Bead size was affected by N source, with the smallest bead occurring with calcium nitrate as N source.
The higher rate of total N applied increased head size but also nearly doubled the incidence of head rot. At the first harvest there was a significant interaction of N rate and source on head rot incidence: head rot increased at the high N rate for five of six N sources, but decreased at the high N rate with sodium nitrate as fertilizer. With potassium nitrate, the tendency for higher rot incidence at the high N rate was reduced (Table 2).
Rate of calcium chloride had no effect on head weight or quality. Leaves of plants grown with potassium or sodium in the fertilizer were noticeably darker green in color.
In the second experiment of 1989, N source had no significant effect on mean head size, although there was a strong trend for heavier heads with urea as N source (Table 3). Yield on an area basis did vary with N source, with urea and potassium nitrate producing the highest yields. Head width varied slightly with N source. Head rot incidence in this trial was extremely low due to the unusually warm, dry conditions during the maturation and harvest period. Rot and downy mildew incidence did not vary with N source. Most favorable bead size occurred with sodium, potassium, and calcium nitrates. Rate of applied calcium had no effect on yield or rot incidence.
Soil pH was affected by N source, with the potassium and sodium fertilizers increasing pH of the surface centimeter of soil and ammonium nitrate and ammonium sulfate reducing pH compared to the control value of 5.4 (Table 4). Use of sodium and potassium nitrates should reduce the amount of lime necessary to maintain adequate pH on heavily fertilized soils.
Leaf tissue N levels were not significantly affected by N source. However, the leaf N levels were lowest with the combination of ammonium sulfate as N source and no applied Ca (data not shown). This may be related to the limitation of N uptake by the low surface soil pH in these plots. Leaf tissue levels of K, S, Ca, Mg, Mn, Fe, and Na were significantly affected by N source but not by rate of applied calcium (Table 4). Ammonium sulfate increased leaf K, S, and Mn concentrations, but reduced levels of Ca, Fe, and Mg. These effects may all be related to the acidifying effect of this fertilizer. The sodium-containing fertilizers dramatically increased leaf Na content.
In the bactericide trial, no rot was observed in either planting on 11 October. On 16 October, rot incidence of 1% was observed on the Erwinia-treated plots in the first field; no rot was observed in the second field. On 20 October, there was moderate to severe phytotoxicity on all Kocide-treated plots in the first field. Severe symptoms included large areas of necrotic floret tissue. Some head rot was also observed, primarily on Kocide-treated plots (Table 5). Severity of the rot was low except for two heads (out of approximately 180 heads total) which exhibited advanced rot of the entire head. Both of these heads were from Kocide-treated plots. No rot was observed in the second field, but mild phytotoxicity was observed on five of the six Kocide-treated plots. The symptom observed was a darkening and hardening of the florets.
On 26 October, a few heads had been cut for commercial harvest in the first field, including a few previously observed to have rot. Of the remaining heads, rot incidence was insignificant except for the Kocide- and WAR60-treated plots. As on 20 October, severe rot occurred only on heads injured by the Kocide treatment. No significant rot was observed in the second field. However, most heads of harvestable size had already been cut. Head rot incidence in these two fields was too low to adequately judge the effect of the antagonists. It is apparent, however, that 4.0 pounds per acre of Kocide is too high a rate to use in attempts to control head rot.
In all three 1990 experiments the source of N had no statistically significant effect on broccoli production, whether expressed on a yield per acre or mean head weight basis. In addition, there were no significant interactions of source and N rate on yield. Thus, only main effects are shown in the tables. No head rot developed in 1990. Application of calcium nitrate had no effect on yield in the first experiment of 1990.
In the first experiment, N rate had a small but significant effect on broccoli yield (Table 6). Both mean head weight and yield on an area basis increased with increasing rate of nitrogen. However, there was very little increase beyond 175 pounds N per acre. There was a trend toward increased bead size of broccoli heads with increasing rate of nitrogen, but the effect was not significant (data not shown).
In the second experiment of 1990, yield and mean head weight increased with increasing rate of N to a maximum at 250 pounds N per acre (Table 7). Yields at 325 pounds N tended to be reduced slightly. High rates of N favored early maturity and larger heads at the first harvest (Table 8). At the second and third harvests of these plots, the largest head size was achieved at less than the maximum rate of N. The effect of applying all N at planting versus delaying the bulk of the application of N until the plants are well-established is seen in Table 9. Even though maximum N uptake does not occur until near head formation, it appears to be critical to provide ample N at time of planting. For a spring planting, this would increase the risk of leaching the N out of the root zone.
In the third experiment, yield on an area basis did not vary significantly with N rate because of variation in stand (Table 10). However, mean head weight varied significantly with N rate, with a maximum at 250 pounds N per acre.
From these trials it appears that amounts of N in excess of 250 pounds per acre exceed crop needs and needlessly increase the chance of nitrate pollution of groundwater or runoff. Nitrogen source has little or no effect on broccoli yield at the rates of N needed for adequate yield and quality.
Table 1. Main effects of N source, N rate, and Ca rate on yield and quality of broccoli, NWREC, Oregon, July, 1989 Mean head Head % good % head % Beadz Treatment wt. (g) width (in.) heads rot mildew size N source NH4NO3 162 3.9 92.7 8.1 1.5 3.6 Ca(NO3)2 171 4.0 90.6 8.3 2.5 3.2 KNO3 164 4.0 90.6 4.5 2.5 3.7 NaNO3 168 4.0 92.6 5.4 1.5 3.6 K/NaNO3 167 3.9 91.1 7.1 1.5 3.5 Urea 169 4.1 93.0 6.8 1.0 3.9 LSD(0.05) NSy NS NS NS NS 0.3 N rate, lb/A 150 160 3.9 93.4 4.8 1.8 3.6 200 174 4.0 90.8 8.6 1.9 3.6 ** ** NS ** NS NS Ca rate, lb/A 0 173 4.1 91.1 7.8 2.1 3.6 10 161 3.9 93.2 5.9 1.7 3.5 20 166 4.0 91.9 6.5 1.7 3.6 LSD (0.05) 8 0.1 NS NS NS NS Notes: No significant differences in stem color or hollow stem incidence. zBead size rated on a five point scale, with 1=very fine bead, 5=very open bead, some open flowers. yNS,*,**: No significant differences, significant at the 5% and 1% levels, respectively. Table 2. Interaction of N source and N rate on broccoli head rot incidence, first harvest, July, 1989, NWREC, Oregon N source N rate (lb/A) % head rot NH4NO3 150 1.9 200 13.9 Ca(NO3)2 150 6.3 200 9.0 KNO3 150 3.4 200 4.8 NaNO3 150 6.0 200 3.9 K/NaNO3 150 4.3 200 6.6 Urea 150 3.5 200 8.8 LSD (0.05) 5.9 Table 3. Main effects of N source and Ca rate on yield and quality of broccoli, September, 1989, NWREC, Oregon Yield Mean head Head % good % head % Bead Treatment (T/A) wt. (g) width (in.) heads rot mildew size N source NH4NO3 4.8 166 3.6 96.1 1.3 2.6 3.3 (NH4)2SO4 4.6 161 3.6 88.7 0.0 11.3 3.3 Ca(NO3)2 4.9 167 3.5 90.4 1.7 7.9 3.1 KNO3 5.2 176 3.7 92.4 0.6 7.0 3.0 NaNO3 4.6 162 3.6 89.9 0.5 9.6 2.9 K/NaNO3 5.0 173 3.8 89.7 1.6 8.6 3.1 Urea 5.8 184 3.8 85.8 0.4 13.8 3.4 LSD(0.05) 0.5 NSz 0.2 NS NS NS 0.3 Ca rate, lb/A 0 5.0 174 3.7 88.4 1.2 10.4 3.3 10 5.0 166 3.6 90.8 1.1 8.1 3.0 20 4.9 170 3.7 92.1 0.4 7.6 3.2 LSD(0.05) NS NS NS NS NS NS 0.2 zNS: No significant differences among means within the column. Table 4. Main effects of N source and Ca rate on soil pH and broccoli leaf elemental concentrations, September, 1989, NWREC, Oregon Percent dry weight Ppm dry weight Soilz Treatment N K S Ca Mg Mn Fe Na pH N source NH4NO3 5.4 2.8 1.0 2.3 0.26 69 190 727 4.8 (NH4)2SO4 5.4 3.7 3.2 1.6 0.18 93 143 603 4.5 Ca(NO3)2 5.8 2.5 1.1 2.4 0.26 62 174 1289 5.4 KNO3 5.7 2.8 1.2 2.5 0.27 56 195 1301 6.0 NaNO3 6.2 2.7 1.1 2.2 0.24 62 175 5698 6.2 K/NaNO3 5.6 2.5 1.1 2.4 0.24 56 175 5330 6.1 Urea 5.9 2.9 0.9 2.0 0.24 62 181 1079 5.4 LSD (0.05) NSy 0.5 0.5 0.3 0.03 17 28 1473 0.3 Ca rate, lb/A 0 5.7 2.8 1.4 2.2 0.23 64 177 2312 5.5 20 5.7 2.9 1.3 2.3 0.24 67 175 2267 5.4 NS NS NS NS NS NS NS NS NS zpH of the surface half-inch of soil at harvest. Unfertilized soil has a pH of 5.4. yNS: No significant differences among means within the column. Table 5. Rot incidence (percent of heads affected) in the first grower field experiment, October, 1989, Marion County, Oregon Treatment Rate 20 October 26 October Untreated control -- 1.0% 0.5% Erwinia carotovora 106 cells/ml 2.0 0.0 Kocide 4.0 lb/A 4.5 14.3 3832 107 cells/ml 0.0 1.0 3871 107 cells/ml 0.0 1.0 WAR60 107 cells/ml 2.2 5.0 LSD (0.05) 2.4 4.7 Table 6. Main effects of N rate and source on yield and mean head weight of 'Gem' broccoli, spring, 1990, NWREC, Oregon Treatment Mean head wt. (g) Yield (tons/acre) N rate (lb/acre) 100 169 5.3 175 175 5.6 250 176 5.7 *z * N source Ammonium nitrate 173 5.2 Calcium nitrate 179 5.6 Potassium nitrate 170 5.4 Sodium nitrate 174 5.6 Urea 172 5.5 NS NS z*,NS: significant at the 5% level and nonsignificant, respectively. Table 7. Main effects of nitrogen rate and source on yield of 'Gem' broccoli for the sum of three harvests, early autumn, 1990, NWREC, Oregon Treatment Yield (tons/acre) Mean head wt. (g) N rate (lb/acre) 100 4.0 167 175 5.0 179 250 5.2 197 325 5.0 187 *z * N source Urea 5.1 184 Sodium nitrate 4.5 181 NS NS z*,NS: significant at 5% level and nonsignificant, respectively. Table 8. Main effects of nitrogen rate and source on mean head weight of 'Gem' broccoli over three harvests, early autumn, 1990, NWREC, Oregon Treatment Mean head weight (g) Harvest 1 Harvest 2 Harvest 3 N rate (lb/acre) 100 164 152 163 175 188 169 179 250 187 192 192 325 204 190 183 * * * N source Urea 179 175 182 Sodium nitrate 192 176 177 NS NS NS Table 9. Effect of splitz versus single application of urea at 250 pounds per acre on mean head weight of 'Gem' broccoli, 1990, NWREC, Oregon 50 pounds at planting 188 g 250 pounds at planting 207 g * zSplit application: 50 pounds/acre at planting and the remaining 200 pounds/acre applied four weeks later after final thinning. Table 10. Main effects of N rate and source on yield, mean head weight and quality of 'Gem' broccoli, late autumn, 1990, NWREC, Oregon Treatment Mean head Yield Bead sizez Hollow wt. (g) (tons/acre) stem (%) N rate (lb/acre) 100 174 4.7 3.0 12 175 209 4.4 2.9 18 250 188 4.8 2.9 19 * NS NS NS N source Potassium nitrate 186 4.2 3.1 13 Sodium nitrate 199 4.8 2.9 16 Urea 184 5.1 2.8 20 NS NS NS NS zFive point scale, with 1=fine and tight, 5=loose, near anthesis.