Research report from OSU's North Willamette Agricultural Research and Extension Center
Delbert Hemphill
OSU Dept of Horticulture, NWREC
Introduction
Nitrate pollution of groundwater resulting from application of high rates of N (up to 300 pounds per acre) to vegetable crops is an increasing concern in the Willamette Valley. These high rates of application exceed crop uptake and leave significant amounts of residual mineral N to be leached during the winter period of high rainfall. Little information is available on N cycling as related to crop rotations or on the ability of winter annual cover crops to capture residual N following the crop.
Partly in response to these concerns, a long-term crop rotation study was established at NWREC in 1988. The planned rotations included grass seed only, grass seed/clover, vegetable/small grains, vegetable/small grains/clover, and vegetable/winter cover crop. The vegetable rotations started with winter wheat, followed by either winter fallow, cereal rye, or cereal rye plus Austrian winter pea cover crop, followed by sweet corn at three N rates in 1990. Following the sweet corn crop, the plots were again fallowed or planted to the same cover crops as in the previous winter. In 1991, these plots were transplanted to broccoli. In addition, plots that had been in red clover since 1989 were also transplanted to broccoli. The primary objectives were to determine how much nitrogen was trapped in the winter cover crops and made available to the broccoli crop (as a function of applied fertilizer N) and to determine the potential of a winter cover legume, or a legume in the crop rotation, to provide N for the broccoli crop. A secondary objective was to determine the effect of weed control by cultivation alone, in the production of transplanted broccoli.
The same plots that had been in winter cover crops in 1990-91 and planted to broccoli in 1991 were replanted to the same cover crops following the broccoli crop. These cover crops were followed in 1992 with a planting of sweet corn. As with the broccoli, the primary objective was to determine the cover crop contribution to the yield of the sweet corn.
Methods
In 1991, all cover crop plots were sampled for biomass and N accumulation on 15 April. Samples were oven-dried, weighed, and subsamples taken for determination of total N. The 30 x 60-foot plots were mowed and disked on 19 April and then plowed, disked, and harrowed in May. On 31 May, all plots received a broadcast, incorporated application of chlorpyrifos at 1.0 pounds/acre and boron at 2.0 pounds/acre. Appropriate "high input" plots also received trifluralin herbicide at 0.75 pounds/acre. On 4 and 5 June, the plots were transplanted to Gem broccoli. Spacing was 12 inches in the row with paired rows on 20-inch centers and a 40-inch wheel track. For plots that had been winter-fallowed or in rye, or rye plus pea cover crops, urea was applied at 125 pounds N/acre to the appropriate subplots on 7 June. An additional 125 pounds N/acre was applied to the appropriate subplots on 1 July.
Four additional plots had been in 'Kenland' red clover. These were split with half of each plot plowed after seed harvest in the fall of 1990 and the remaining half plowed in April, 1991, permitting substantial shoot regrowth. These subplots were further split by application of either 0 or 200 pounds N/acre as urea. The N application was split, with half the total applied on 7 June, and the remainder on 1 July.
All plots were sprinkler-irrigated for four weeks, after which drip irrigation tubing was installed between the paired rows. The tubing had emitters at 9-inch intervals and a flow rate of 0.5 gallons/minute/100 feet. All plots were tractor-cultivated on 24 June and hand-hoed in July. Leaves were sampled for N content determination on 17 July. All plots were harvested on 30 July and 7 August. Yields were determined from a 40 row-foot section near the center of each subplot.
After disking and harrowing, cover crops were seeded on 20 September. Eight of the 16 plots were planted to 'Wheeler' cereal rye at 65 pounds/acre. The other eight plots were planted to a mixture of 'Wheeler' rye at 35 pounds/acre and Austrian winter pea at 100 pounds/acre. No fertilizers or pesticides were applied to the cover crops. Nitrogen rate subplots of 600 square feet each were determined by the N rate applied to the broccoli crop.
On 7 April, 1992, samples were taken from all subplots for determination of shoot dry weight and N uptake. The shoots were clipped about 1 inch above ground. The rye and pea plants were counted, weighed, and analyzed separately. The cover crops were mowed down on 17 April, disked on 27 April, and plowed under on 28 April, 1992. The plots were disked and harrowed in early May.
On 19 May, 1992, 2.0 pounds EPTC/acre was applied to one rye and one rye-pea mixture plot in each of the four blocks ("low input" or reduced herbicide main plots). The EPTC was rototilled into the surface 3 inches of soil. 'Jubilee' sweet corn was seeded in 20-inch paired rows on 20 May. The distance between pairs of rows was 40 inches. Triple superphosphate was banded 2 inches to the side and 2 inches beneath the seed row at a rate of 80 pounds/acre. Immediately after planting, the remaining rye and rye-pea plot in each block was sprayed with 2.0 pounds atrazine and 3.0 pounds alachlor/acre ("high input" or high herbicide main plots).
On 21 May, one-half of the total N was applied as urea in a surface band between the paired rows at rates of 0, 50, and 200 pounds/acre. These N rate subplots were in the same location as the corresponding N treatments on the previous vegetable crops. Drip irrigation tubing was then installed between each pair of rows. Two weeks after emergence, the corn was thinned to a stand of 7 inches between plants in the row. The remainder of the urea was sidedressed on 3 July.
On 3 July, the "low input" main plots that had been in cereal rye were overseeded with 'Wheeler' cereal rye at 50 pounds/acre. The "low input" main plots that had been in rye plus pea were overseeded with 'Kenland' red clover. In each case, the seed was broadcast with a spinner-type fertilizer spreader and scratched in with a garden rake. The overseeding was in preparation for the 1992-1993 experiments, in which one of the objectives will be to determine the value and feasibility of overseeding a cover crop into the standing vegetable crop.
On August 16, the stalks from 15 feet of two inner rows of each suplot were harvested. Ears were counted, measured for length, and rated for tipfill. Subsamples of ears and stalks were taken for determination of dry weight and total nitrogen content.
Results and Discussion
Cover Crop N Recovery, 1991
Nitrogen added to the sweet corn crop in the summer of 1990 significantly affected the growth of the cover crops in the winter of 1990-91 (Figure 1). Both the cover crop biomass and N uptake (Figure 2) increased as the rate of N applied increased. To estimate the amount of N recovered from the fertilizer applied to the sweet corn, the amount of N recovered in the cereal rye without applied N can be subtracted from the amount recovered with 200 pounds/acre applied N. This difference of about 80 pounds N/acre suggests that, without a cover crop, 80 pounds/acre would have been available for leaching.
Adding Austrian pea to the cover crop increased N content of the total cover crop at the 0 and 50 pound N rates, but not at the 200 pound rate. The legume contributed only about 10 pounds N/acre, but there was a synergistic effect on the companion rye crop, which accumulated additional N in the presence of the legume. Legume growth was suppressed at the high rate of N, probably by competition with the vigorously growing rye. The number of pea plants per unit area decreased from 30-35 per square meter at 0 or 50 pounds applied N per acre to 11 per square meter at 200 pounds N/acre. Thus, adding legumes to a cover crop mix in order to fix N may be effective only following low rates of N application to the preceding crop.
Broccoli Response to the Preceding Cover Crop, 1991
With transplanting allowing the crop a head start on weed growth, and with the subsequent cultivation of all plots, herbicide application had no effect on yield (Table 1). This suggests that it is practical to grow transplanted broccoli with only mechanical tillage and obtain satisfactory yields.
The cereal rye cover crop failed to increase broccoli yield, even though it trapped a significant amount of nitrogen from the previous corn crop (Table 1). The yield of plots with a cereal rye cover and no herbicide was actually reduced significantly compared to the fallow plots. The rye plus pea cover crop also failed to significantly increase broccoli yield. This is in contrast to the 1990 sweet corn crop, where the combination of rye and Austrian winter pea increased yield, particularly at low rates of applied nitrogen. One can speculate that the cereal rye had an allelopathic effect on the growth of the broccoli root system or that decomposition of the cover crop took applied N from the broccoli crop rather than acting as a source of readily available N.
For the plots that had been in red clover, yields were generally higher than for the fallowed plots or the winter-cover crop plots (Table 2). Plots receiving applied N produced greater broccoli yields than those not receiving applied N, regardless of whether the cover was plowed in the spring or fall. Spring-plowing was definitely advantageous, as yields exceeded those with fall-plowing. The yield of broccoli following clover was greater, with no applied N, than for broccoli following winter fallow or winter cover crops. The greatest yield in this experiment (4.6 tons/acre) was with the combination of spring-plowed cover and 200 pounds N/acre.
The disorder hollow stem is, at least in part, related to high rates of N and rapid growth. In this trial, the incidence of hollow stem increased with applied N compared to no applied N, with pea plus rye compared to rye only as cover crop, and with spring rather than fall plowing of the preceding clover crop. In each case, this is consistent with hollow stem being related to high rates of available N.
Table 1. Main effects of winter cover crop and N rate on yield, head size, and incidence of hollow stem of transplanted 'Gem' broccoli, NWREC crop rotation study, 1991 First harvest Sum of two harvests Yield Head wt. Hollow stem Yield Head wt. Hollow stem (T/A) (g) (%) (T/A) (g) (%) Cover crop Fallow 2.1 160 40.7 2.8 151 34.4 rye, - herb. 1.2 123 14.9 2.2 122 11.7 rye+pea, -herb. 2.1 156 35.0 3.1 155 28.5 rye, +herb. 1.8 136 19.8 2.7 134 16.3 rye+pea, +herb. 1.9 143 44.4 2.8 143 33.0 LSD (0.05) 0.4 18 16.9 0.4 20 12.3 Contrasts: Fallow vs. others NS * * NS NS * -herb. vs. +herb NS NS NS NS NS NS rye vs. rye+pea * ** ** * ** ** N rate (lb/A) 0 1.1 110 14.6 2.1 109 11.4 125 2.7 170 66.5 3.1 160 48.0 250 2.9 197 58.4 3.8 196 49.7 Significance Q** Q** Q** L** Q** Q** NS,*,**,L,Q: No significant differences, differences significant at 5%, 1% levels, respectively, linear, quadratic. Table 2. Main effects of fall versus spring plowing of an established red clover seed crop and nitrogen rate on yield, head size, and incidence of hollow stem of transplanted 'Gem' broccoli, NWREC crop rotation study, 1991 First harvest Sum of two harvests Yield Head wt. Hollow stem Yield Head wt. Hollow stem (T/A) (g) (%) (T/A) (g) (%) Plowing season Fall 2.3 180 47.7 3.3 176 37.5 Spring 3.5 193 65.3 3.9 186 58.4 Significance ** * * * * * N rate (lb/A) 0 2.0 143 43.0 2.8 140 33.4 200 3.6 224 70.0 4.2 222 62.6 Significance ** ** ** ** ** ** *,**: Significant at 5% and 1% levels, respectively.
Cover Crop N Recovery, 1992
Both increasing the fertilizer rate on the preceding broccoli crop and the presence of peas in the cover crop increased total cover crop yield and N uptake (Figures 3 and 4). The high or reduced herbicide inputs on the broccoli had no effect on cover crop yield or N uptake, so the results are averaged over herbicide treatment. The rye produced greater biomass in the presence of peas at the low and intermediate rates of residual N. Pea biomass was somewhat reduced at the highest rate of N. The peas did not have as much effect on N uptake by the rye. Pea N uptake was nearly constant over N rates.
A rough estimate of the amount of residual fertilizer N left over from the broccoli crop and recovered by the rye cover crop can be obtained by examining the rye-only uptake at the three fertilizer rates shown in Fig. 2. Subtracting the amount of N taken up by the rye grown on non-fertilized subplots from the N taken up at the other two N rates suggests that about 9 pounds N/acre were taken up from the intermediate rate of N, and 22 pounds from the high rate of N. This N would have been available for leaching. Of course, an undetermined amount of N may have leached before the cover crops were well established.
For the high N rate subplots, the N uptake by the cover crop increases from 41 pounds/acre for rye alone to 76 pounds/acre for rye plus pea. A large part of the extra N taken up by the rye plus pea may come from N fixation, but the peas may also have taken up some residual N that escaped the rye.
The population density of Austrian winter pea plants in the rye plus pea plots was significantly lower in the high herbicide treatment than in the reduced herbicide treatment (Figure 5). Interestingly, the herbicide used for the "high input" treatment is registered for use on dry peas and is generally considered safe for the crop.
At corn harvest, pickers were instructed to pick all ears that they judged to have any mature kernels. The herbicide program (or overseeding to rye or clover) had no effect on the number of ears harvested (Table 3). However, both increasing rate of applied N and the presence of Austrian peas in the preceding cover crop increased the number of ears judged harvestable. Likewise, ear yield (tons/acre) did not vary with the herbicide program but did increase with increasing N rate and pea in the cover crop. There were no significant interactions among herbicide program, cover crop, and N rate affecting any component of yield or tipfill; thus, only main effects are given in Table 3. The highest yield of 11.0 tons/acre was for plots with the combination of atrazine-alachlor as herbicide (no overseeding), the highest N rate, and the mixture of rye and pea as cover crop.
Ear length was not affected by cover crop or herbicide program, but mean ear weight was affected by both cover crop and herbicide, as well as N rate. Kernel moisture content was not affected by treatment, while stover (stalk) weight increased with increasing rate of N. Tipfill, a measure of kernel development at the silk end of the ear, was increased by the atrazine-alachlor herbicide program as well as by increasing the rate of applied N.
The most striking feature of these results is the contribution of the cover crop peas to the sweet corn yield, even at a high rate of applied N. It is interesting to note that the 10 tons/acre yield achieved with 50 pounds N/acre following a rye plus pea cover crop is equivalent to the yield with 200 pounds applied N/acre when the cover crop was rye only (Figure 6). In this experiment it was not possible to measure the contribution of the rye-only cover crop to sweet corn yield, as the program of crop rotations did not allow for sweet corn following a winter fallow.
Table 3. Main effects of herbicide and overseeding, preceding cover crop, and rate of applied nitrogen on ear yield and quality of sweet corn, NWREC, 1992 Treatment No. Yield Length Mean Tipfillz Kernel Stover harvested/ (Tons/ (inches) wt. moisture yield acre acre) (g) content (%) (T/A) Herbicide, overseeding EPTC, overseeded 30,100 8.7 11.1 260 2.8 77.8 14.6 Atrazine, not overseeded 29,210 9.1 11.2 279 3.1 75.7 14.7 Significance NS NS NS * * NS NS Cover crop Rye 27,780 8.1 11.2 258 2.9 77.2 14.0 Rye + pea 31,530 9.7 11.2 281 3.0 76.3 15.3 Significance * * NS * NS NS NS N rate (lb/A) 0 25,110 7.2 10.9 251 2.6 77.7 12.8 50 30,200 9.2 11.2 276 3.1 76.4 14.6 200 33,650 10.3 11.4 280 3.3 76.2 16.6 Significance ** ** ** ** ** NS ** zRated on a 5-point scale with 5 = perfect kernel development to the tip of the ear; 1 = at least two inches of undeveloped or shriveled kernels.