Brassica Green Manure Systems for Weed, Nematode, and Disease Control in Potatoes

Objectives 1, 2, and 4. Brassica ground cover. 'Humus' winter rape, which is winterhardy, provided 93 to 100% ground cover from early to mid-October until incorporation the following spring. 'Martigena' white mustard, which is not winterhardy, provided excellent ground cover (>95%) after establishment until winter kill in mid-to late October, 55 to 66% cover in mid-December, and 43 to 60% cover the following spring before planting potatoes. Thus, both green manures provided good to excellent fall, winter, and early spring soil erosion protection.

Brassica biomass and glucosinolate production. Total dry weight just before winter rape incorporation at Aberdeen ranged from 8170 to 10820 kg/ha, depending on year, and total dry weight just before white mustard winter kill ranged from 5550 to 5720 kg/ha. At Prosser, winter rape dry weight ranged from 10700 to 15200 kg/ha just prior to winter rape incorporation and white mustard dry weight ranged from 5080 to 6320 kg/ha just before winter kill. Glucosinolate profiles varied between the two green manures. Benzyl glucosinolate and 4-hydroxy benzyl glucosinolate were the major glucosinolates in both roots and shoots of white mustard at Aberdeen. Major shoot glucosinolates in winter rape were 4-pentenyl, 2-hydroxy-3-butenyl, and 2-hydroxy-4-pentenyl (80% of total shoot glucosinolates); major root glucosinolates were phenylethyl, 4-pentenyl, and 2-hydroxy-3-butenyl (76% of total root glucosinolates). Total glucosinolate concentrations were higher in white mustard than in winter rape, but winter rape produced more biomass than white mustard, which resulted in greater total glucosinolate production (biomass x concentration) for winter rape than for white mustard.

Weeds. Redroot pigweed, hairy nightshade, and common lambsquarters were the major weeds present at Aberdeen and common lambsquarters and redroot pigweed were the major weeds infesting the Prosser experimental site. Average weed populations at row closure in early July in the no green manure, low input (no herbicide) treatment at Aberdeen were 45 redroot pigweed/m², 19 hairy nightshade/m², and 12 common lambsquarters/m². At Prosser, weed populations in mid-June were 31 common lambsquarters/m² and 21 redroot pigweed/m² in the no green manure, low input treatment.

To evaluate the effect of green manures alone on weed control, weed biomass in the low input winter rape and white mustard treatments was compared to biomass in the no green manure, low input treatment. At Aberdeen, late season hairy nightshade biomass was reduced by winter rape or white mustard incorporation in 1996, but common lambsquarters and redroot pigweed biomass were not. In 1997, early season hairy nightshade and total weed biomass were lower in the winter rape treatment than in the white mustard or no green manure treatment. However, surviving plants grew well and by the end of the season there was no effect of green manure incorporation on hairy nightshade or total weed biomass. At Prosser, common lambsquarters made up 91% of the late season total weed biomass. Winter rape incorporation reduced total weed biomass 60% and common lambsquarters biomass 78% compared to the no green manure, low input treatment. Redroot pigweed biomass was not reduced by either winter rape or white mustard incorporation. Although winter rape incorporation substantially reduced common lambsquarters biomass, control was not commercially acceptable.

Winter rape provided some weed control at Prosser, but provided little or no weed control at Aberdeen. Differences in performance may have been due in part to soil or environmental conditions. The soil type at Prosser was a loamy sand with 0.88% organic matter, which would have less adsorptive capacity, and therefore, less tie-up of glucosinolate degradation products than the silt loam with 1.3% organic matter at Aberdeen. Soil pH was lower at Prosser than Aberdeen, which may have affected type and rate of formation of glucosinolate degradation products. Soil temperatures also were lower at Aberdeen than at Prosser, which may have affected both the rate of glucosinolate degradation and the rate of release of degradation products from incorporated tissues.

Management levels had a highly significant effect on weed biomass at both Aberdeen and Prosser. The medium and high input treatments reduced late-season (September) common lambsquarters, hairy nightshade, and redroot pigweed biomass 98 to 100% at Aberdeen. At Prosser, late season common lambsquarters and redroot pigweed biomass was reduced 96 to 99%. The combination of green manure incorporation plus a low-rate, postemergence application of rimsulfuron + metribuzin provided weed control equal to the high input treatment at both Aberdeen and Prosser.

Rhizoctonia infection. Both below ground stem infection and tuber infection by Rhizoctonia solani were evaluated at Aberdeen. A higher percent of below ground stems (56%) had less than or equal to 5% Rhizoctonia infection in the winter rape green manure treatment than in the white mustard (37%) or no green manure (35%) treatment. Management level did not significantly affect the percent of clean stems (less than or equal to 5% rhizoctonia infection), although there was a trend for more clean stems in the high input system than in the medium or low input systems.

Green manure effect on tuber infection by Rhizoctonia varied with management level. In the unfumigated low and medium input treatments, there was a higher percent of tubers with less than or equal to 5% of the surface covered with Rhizoctonia when winter rape was incorporated (64%) than when white mustard (27%) or no green manure (28%) was incorporated. However, in the fumigated, high input treatment, there was no difference in tuber infection between winter rape and no green manure subplots because fumigation decreased tuber infection in the no green manure treatment.

Verticillium wilt. Visual evaluation of Verticillium wilt showed significantly less wilt incidence in the winter rape treatment than in the white mustard or no green manure treatment. In 1997, potato stems in the winter rape green manure treatment also were less colonized by Verticillium dahliae than potato stems in the white mustard or no green manure treatments. However, V. dahliae populations in the soil were not different among green manure treatments, suggesting that winter rape incorporation may not have a direct toxic effect on V. dahliae, but may reduce potato root infection by the disease. Additional studies to determine winter rape green manure effects on potato root infection by V. dahliae should be conducted. Management level affected soil populations of V. dahliae; populations were lower (P=0.07) in the fumigated, high input treatment than in the unfumigated low and medium input treatments. There was no significant effect of management level, averaged over green manure treatments, on wilt incidence or stem colonization, although there was a trend for less wilt incidence in the high input management system than in the medium or low input systems.

Root lesion nematodes. Root lesion nematodes (Pratylenchus neglectus) were present at the Aberdeen site each year of the study. Soil samples taken just before green manure incorporation in unfumigated plots showed that root lesion nematode populations in the winter rape plots were 76% lower than in the white mustard and no green manure plots. However, root lesion nematode populations were 3.8 times higher in the unfumigated winter rape treatment than in the unfumigated white mustard or no green manure treatment after green manure incorporation. Because the root lesion nematode is a migrating endoparasitic nematode, it multiplied inside winter rape roots; when roots began to decay after winter rape incorporation the developing young along with adults were released into the soil. Although root lesion nematodes can infest potatoes, post-incorporation populations in the unfumigated winter rape treatment (1450 nematodes/500 cm³) were below the economic threshold of 2000 nematodes per 500 cm³ of soil. However, in soils with higher initial root lesion nematode populations than we had in our study, winter rape may be unsuitable as a green manure because root lesion nematode populations increase in its presence.

Tuber biomass. Tuber biomass in the winter rape treatment was significantly less than in the white mustard and no green manure treatments at all sampling dates both years of the study at Aberdeen and at the mid and late season sampling dates at Prosser. Based on soil tests conducted at Aberdeen we hypothesize that lower tuber biomass in the winter rape treatment may have been the result of nitrogen mineralization from winter rape incorporation. Both years, inorganic nitrogen availability was higher in the top 30 cm of the soil profile during tuber initiation in the winter rape treatment than in other treatments, which may have delayed tuber initiation and development.

Management level had no effect on tuber biomass at the early and mid season sampling dates at either Aberdeen or Prosser, but at the late season sampling date, tuber biomass was lower in the low input management treatment than in the medium or high input treatments. This likely was the result of poor weed control in the low input treatment compared to the other treatments.

Potato tuber yield. U.S. No. 1 and total tuber yield at the Aberdeen location responded differently to treatments in 1996 compared to 1997, so yield data were analyzed by year. Management level had a significant effect on yield both years. In 1996, U.S. No. 1 tuber yield in the low input management system was 17 and 28% percent lower than in the medium and high input treatments, respectively, and total tuber yield in the low input treatment was 15 and 17% lower than in medium and high input treatments, respectively. In 1997, U.S. No. 1 yield in the low input treatment was 38 and 45% lower than in the medium and high input treatments, respectively, and total yield in the low input treatment was 25 and 29% less than in the medium and high input treatments, respectively. Green manure treatments also had significant effects on tuber yields. In 1996, U.S. No. 1 and total tuber yield were 43 and 26% lower, respectively, in the winter rape green manure treatment than in the no green manure treatment. Within the U.S. No. 1 category, yield of tubers in all weight categories was lower in winter rape green manure treatment than in the white mustard and no green manure treatments. White mustard incorporation did not affect either U.S. No. 1 or total tuber yield. In 1997, the winter rape green manure treatment did not affect U.S. No. 1 tuber yield, but reduced total tuber yield 10% compared to the no green manure treatment. The white mustard treatment had no effect on either U.S. No. 1 or total tuber yield. Within the U.S. No. 1 category, yield of tubers in the 113 to 170 g weight category was lower for the winter rape green manure treatment than for the white mustard and no green manure treatments. However, yield of tubers in the greater than or equal to 340 g weight category was higher for the winter rape green manure treatment than for the white mustard and no green manure treatments.

Reduced yields in the winter rape treatment at Aberdeen in 1996 may have been the result of untimely nitrogen mineralization from winter rape incorporation. Inorganic nitrogen levels were higher from mid-June to mid-July in 1996 than in 1997, which exacerbated effects on tuber development and contributed to the reduced yield observed in 1996.

U.S. No. 1 yields at Prosser were affected by management level (P=0.07) but not by green manure treatment. U.S. No. 1 yields in the low input management system were 13 and 10% lower than in the medium and high input systems, respectively. Green manure treatment did not affect overall U.S. No. 1 yield but did alter the distribution of tubers within the U.S. No. 1 weight categories in a manner similar to that observed at Aberdeen in 1997. Tuber yield in the 113 to 170 g weight category was less in the winter rape green manure treatment than in white mustard or no green manure treatment, but tuber yield in the greater than or equal to 340 g weight category was greater in the winter rape green manure treatment than in the white mustard or no green manure treatment. Neither management level nor green manure treatment significantly affected total tuber yield at Prosser.

Conclusions. Both winter rape and white mustard provided good to excellent fall, winter, and early spring ground cover, but winter rape provided better weed and disease suppression than white mustard. The winter rape green manure consistently reduced common lambsquarters biomass at Prosser but did not consistently suppress weeds at Aberdeen. However, a combination of green manure incorporation plus a low rate, postemergence herbicide application provided weed control similar to the high input, no green manure, standard practice treatment at both locations. Winter rape incorporation also reduced Rhizoctonia and Verticillium wilt in potatoes compared to the no green manure control at Aberdeen.

Nitrogen in winter rape roots and shoots was mineralized after incorporation, resulting in elevated soil nitrate levels in the top 30 cm of soil during potato tuber initiation. Delayed tuber initiation was observed at both Aberdeen and Prosser. U.S. No. 1 yields were not reduced by green manure treatments at Prosser, but were reduced one out of two years at Aberdeen. The longer growing season at Prosser may allow time for full recovery from nitrogen-delayed tuber initiation, while the shorter season at Aberdeen increases the potential for nitrogen-delayed tuber initiation to affect final tuber yield. Until nitrogen management practices for a winter rape system for shorter growing season conditions have been developed and tested on-farm, recommendations for green manure systems will differ for shorter vs. longer growing season areas. Both the winter rape and white mustard systems provide good to excellent ground cover compared to standard practices that leave the soil bare, but neither system provides commercially acceptable weed control. However, the winter rape system provides Rhizoctonia and Verticillium wilt suppression, and better weed suppression than the white mustard system. Therefore, to maximize pest suppression and erosion control while minimizing risk to tuber yield in longer growing season potato production areas, a fall-planted, spring incorporated winter rape green manure + low rate, postemergence herbicide system is recommended. In shorter growing season areas, weed and erosion control can be maximized by use of a white mustard + low rate, postemergence herbicide system. Because white mustard is winterkilled, no green tissue is incorporated and untimely nitrogen mineralization is not a concern.

Objective 3. A change in potato production practices can affect revenue by changing yield, grade, quality, and price. Potato growers are not paid on total yield but only for tubers meeting specified size and grade standards. Total yield was converted to paid yield by removing the tubers classified as culls and undersized tubers. Since the USDA tuber size categories used in grading potatoes from the plots do not match the categories used in processing potato contracts, an approximation was made to estimate the undersize tubers. Fifty percent of the under 4 ounce tubers were classified as undersize. This was applied to yield data from all treatments. Yield data for 1996 and 1997 were combined and converted to a paid yield by subtracting tubers classified as culls and 50 percent of the under four-ounce tubers from the total yield. Table 1 contains the paid yield data by year and the average paid yield for each of the treatments.

The price per hundredweight that a potato grower receives is influenced by the quality of the potatoes. An incentive-adjusted price commonly found in contracts with potato processors was used to calculate gross revenue. A base price of $4.50 was adjusted using quality incentives. The quality incentives included: percent No. 1 tubers (+/- 1.5 cents per percentage point above/below 50 percent), a size incentive for 6 to 14 ounce tubers ( +/- 2.5 cents per percentage point above/below 44 percent) and a specific gravity incentive (+/- 2.0 cents per thousandth point above/below 1.078). The incentive-adjusted price for each treatment is shown in Tables 2, 3, and 4.

A baseline budget at each input level (low, medium and high) was run without the green manure. Herbicides and fumigation were the only input levels that were varied directly. Other inputs, such as fuel and labor, varied indirectly in response to changes in machinery operations. The low input level was no herbicide and no fumigant. The medium input level was no fumigant and a herbicide tank mix of 1 ounce of Matrix and 0.167 pounds of Sencor DF. The high input level was fumigation with 50 gallons of Vapam per acre and a herbicide tank mix of 1.5 ounce of Matrix and 0.5 pounds of Sencor DF per acre. The Sencor DF price was $23.35 per pound and the Matrix price was $14.80 per ounce. Vapam was $3.20 per gallon and the custom application charge was $30 per acre.

Green manures were fall-seeded at 7.0 pounds per acre. The price of winter rape was $1.25 per pound and the price of white mustard and brown mustard was $0.85 per pound. Green manures were mechanically incorporated in the spring prior to planting potatoes.

Tables 2, 3, and 4 show the abbreviated costs and returns estimates for the low input, medium input and high input plots, respectively. Each table shows the revenue and costs of the three green manure systems. Comparisons between tables show how input levels influence revenue and costs, while comparisons within the same table show how the green manure systems affected revenue and costs. Comparisons can also be made as to how input level and green manure systems affected paid yield and the quality of the crop by examining the incentive-adjusted price.

At the low input level, the plots with no green manure had the highest paid yield. The white mustard green manure system produced a higher quality crop when compared to the system with no green manure as evidenced by the higher incentive-adjusted price. The higher price was not sufficient, however, to compensate for the lower paid yields. The plots with no green manure produced the highest gross revenue. The system with no green manure had the lowest operating costs as it did not incur the additional costs associated with the green manure crop. With higher revenue and lower costs, the plots with no green manure produced the highest net return (smallest negative return).

At the medium input level, low-rate herbicide application brought higher gross revenue because both paid yield and incentive-adjusted price increased when compared to the low input system. While the added inputs increased operating costs, revenue increased more, resulting higher returns above operating costs and higher net returns. The plots with no green manure still produced the highest net return, although the green manure system using white mustard was a close second.

At the high input level, both paid yield and the incentive-adjusted price increased, resulting in the highest gross revenue of the three input levels. Fumigation increased operating costs by $190 per acre. This was greater than the revenue increase for all but the green manure using winter rape, which resulted in net returns lower than with the medium input level.

Extending plot-level data to farm-based enterprise budgets is a relatively crude method for making economic comparisons. The values generated in this type of analysis are valid only in making relative comparisons within the study alternatives. The validity of any conclusion is tied to the assumptions used in developing the economic analysis, the continued unconstrained availability of inputs, and the continuation of the present price structure.

The low input system with or without a green manure had the lowest net return (highest negative return). The added costs of the green manure system were not offset by higher revenues from either an increase in paid yield or an increase in the incentive-adjusted price, and therefore only increased the already negative return. Both the high input system and the medium input system provided substantially higher net returns compared to the low input system. The medium input systems with white mustard or no green manure produced higher net returns than the high input system with these green manure treatments.