Study site.

A 0.43-ha field of northern highbush blueberry (Vaccinium corymbosum L.) was established in Oct. 2006 on a site in transition to organic production at the North Willamette Research and Extension Center (lat. 45°16′47.55″ N, long. 122°45′21.90″ W), Aurora, OR. Winter wheat (Triticum sp.) was planted in the field for at least 10 years before the study. Soil at the site is a Willamette silt loam (fine-silty mixed superactive mesic Pachic Ultic Argixeroll). Organic matter content in the soil averaged 3.7% at planting, and soil pH averaged 4.9. The planting was certified organic by Oregon Tilth (OTCO, Salem, OR) in May 2008. Additional information on site preparation, planting, weed control, and irrigation is provided in Larco (2010).

Experimental design.

Forty-eight treatments were arranged in a split-split plot design and included two bed types (flat and raised) as main plots, four fertilizer treatments (feather meal and fish emulsion applied at rates of 29 and 57 kg·ha⁻¹ of N) as subplots, and a combination of three mulch treatments (sawdust, compost + sawdust, and weed mat) and two cultivars (Duke and Liberty) as sub-subplots. There were five replicates. The five main plots consisted of two rows of plants each (one flat and one raised), and each sub-subplot consisted of six plants. Rows were spaced 3.0 m apart, and plants were spaced 0.76 m apart within sub-subplots (4385 plants/ha) and 1.5 m apart between sub-subplots. A guard row was also planted on each side of the field.

‘Duke’ and ‘Liberty’ plants were obtained from a local nursery as 2-year-old container stock and transplanted on 9 Oct. 2006. The field was ripped and rototilled before planting. Raised beds (0.3 m high and 1.5 m wide at the base) were formed using a bed shaper. Douglas fir sawdust (Pseudotsuga menziesii M.; Decorative Bark, Lyons, OR) and yard debris compost (Rexius, Inc., Eugene, OR) mulch were applied on 12 Oct. 2006. Beds mulched with sawdust were covered 9 cm deep (360 m³·ha⁻¹) whereas those mulched with compost + sawdust were covered with 4 cm of compost (152 m³·ha⁻¹) followed by 5 cm of sawdust (200 m³·ha⁻¹). We selected yard debris compost for this study because it had relatively low salt (electrical conductivity) and we could apply high rates of stable organic matter without danger of excess plant-available N application (Larco, 2010). The slow-release N provided by the yard debris compost was estimated at roughly 25 kg of N·ha⁻¹/year (3% of total compost-N applied) based on our experience with similar composts applied before planting in grass (Sullivan et al., 2003) and sweet corn (Gale et al., 2006). The compost was topped with sawdust with the goal of creating a barrier to weed establishment. The mulches were spread mechanically in 0.75-m-wide strips on each side of the plant rows. Black weed mat (Baycor; TenCate Protective Fabrics, Union City, GA) made from woven polypropylene was installed on 1 Feb. 2007. The weed mat was 1.5 m wide and centered over the planting beds before securing it in place with landscape staples. A 20-cm-diameter hole was cut in the weed mat for each plant and covered with 5 cm of sawdust mulch (1.4 m³·ha⁻¹) after planting. Certified organic grass seed (Festulolium braunii K. Richt.) was planted between rows at a rate of 28 kg·ha⁻¹ on 14 Oct. 2006.

Feather meal (Nature Safe, Cold Spring, KY; 13N–0P–0K) was applied around the base of the plants in two equal applications on 3 Apr. and 16 May in 2007, 4 Mar. and 22 Apr. in 2008, and 4 Mar. and 15 Apr. in 2009. Fish emulsion (Fish Agra; Northeast Organics, Manchester-by-the-Sea, MA; 4N–0.4P–0.8K) was also applied around the base of the plants but every 2 weeks from 16 Apr. to 9 July in 2007, 15 Apr. to 8 July in 2008, and 8 Apr. to 1 July in 2009. To reduce risk of salt damage to the plants, fish emulsion was diluted with 10 parts water before each application. Plants fertilized with feather meal were deficient in N during the first summer after planting (Larco, 2010) and, therefore, were fertilized with an additional 4 kg·ha⁻¹ of N as fish emulsion on 28 June and 5 July 2007. The plants recovered quickly and did not require additional fertilization with fish emulsion in subsequent years.

Field management.

Plants were irrigated using a single line of polyethylene drip tubing (Netafim, Fresno, CA). The tubing had 2 L·h⁻¹ pressure-compensating, inline emitters spaced every 0.3 m. Lines were located along the row near the base of plants and either placed beneath the weed mat or buried under the organic mulches. Irrigation was controlled by electric solenoid valves and an automatic timer set weekly and scheduled based on changes in soil water content [see Larco (2010) for details]. A second line of drip (located near the base of the plants, but on the other side of the row of the first line) was also installed in weed mat plots to allow for additional irrigation to compensate for increased water use associated with higher temperatures in the treatment (Neilsen et al., 2003; Williamson et al., 2006).

Weeds were removed by hand from plots mulched with sawdust and weed mat and by a combination of 20% acetic acid (vinegar) applications, propane-flaming, and hand-weeding in plots mulched with compost + sawdust. Vinegar applications were done after 1000 hr on days with no rain and light wind in 2007 to 2008 (Webber et al., 2005). A handheld propane flamer was constructed with a 0.20-m-diameter head and an 860,000 to 1,505,000-kg·m⁻¹ small gas heater attached to a rod to facilitate spot burning in 2008. Lemon grass oil (50% a.i.; Green Match EX^®; Marrone Bioinnovations, Davis, CA) was used as a directed post-emergence spray to weeds in compost + sawdust plots in 2009. Hand-pulling or herbicide treatments, depending on mulch treatment, were done only when weed coverage was more than 20%.

Plants were pruned in Oct. 2006 (Year 1) and Jan. 2008 (Year 2) and 2009 (Year 3). To maximize vegetative growth during establishment, all fruit buds were removed from the plants in the first year after planting (Oct. 2006; Strik and Buller, 2005). The next year, pruning was adjusted to plant size with proportionally more fruit buds left on larger plants and few or none left on the smallest plants.

‘Duke’ was harvested from 3 July to 23 July and ‘Liberty’ was harvested from 23 July to 13 Aug. in 2008. In 2009, fruit harvest was from 2 June to 17 July and 22 July to 22 Aug. for ‘Duke’ and ‘Liberty’, respectively. Ripe fruit was picked weekly by hand and weighed to determine the total yield per plot. Fruit were sold to a fresh-market shipper and the price was recorded.

Statistical analysis.

Analysis of all treatment effects on yield and pruning time was done as a complete factorial for a split-split plot design using the PROC MIXED procedure in SAS software package Version 8 (SAS Institute, Cary, NC). Contrasts were used to compare the effect of fertilizer source and rate on net returns.

Economic budget analysis.

Input costs (labor, equipment, and product) were recorded from establishment (Year 0; 2006) through Year 3 (2009) for all activities required to maintain the treatment plots and were extrapolated to a per-hectare cost. Cash cost and return budgets were constructed for each treatment to compare net returns from each production system as well as to evaluate each treatment independently. Cost assumptions used in the assessment of the various treatments were as follows: 1) blueberry plants cost $3.15 each for ‘Duke’ and $3.45 each for ‘Liberty’; 2) pre-plant land preparations (plowing, disking, etc.) were contracted using a custom farming provider at $30/h; 3) weed mat cost $2965/ha and has a life expectancy of 5 years; 4) sawdust mulch cost $7.85/m³, and compost mulch cost $11.35/m³. Sawdust and compost mulches were custom-applied separately. Application of sawdust alone cost $1050/ha, whereas application of both sawdust and compost cost $1482/ha; 5) the drip irrigation system cost $5683/ha and has a life expectancy of 10 years. Repair and maintenance of the system cost 1% of the purchase price per year; 6) the trellis system cost $4200/ha and was amortized over a 30-year life. Repair and maintenance of the system cost 1% of the purchase price per year; 7) expenses to reduce bird depredation were $124/ha per year; 8) seven bee hives/ha were rented during production years at a cost of $50 each; 9) all labor was valued at $15/h, which included worker’s compensation, unemployment insurance, and other labor overhead expenses; and 10) the tractor and sprayer used to apply acetic acid cost $17/h, inclusive of all associated variable and fixed costs. The wage rate of the tractor operator was $15/h.

Bed type.

Raised beds cost $741/ha more than flat beds (Table 3) to establish but increased gross returns by an average of 63% in 2008 to 2009 (Table 2) and cumulative net returns by an average of $2861/ha (Tables 4 and 5). In 2009, returns were similar across mulch types on raised beds but were greater for compost + sawdust and weed mat mulches when grown on flat ground (Table 2). Plant growth and subsequent fruit production were greater on raised beds than on flat ground likely as a result of improved soil drainage and/or aeration (Larco, 2010; Scherm and Krewer, 2008).

Table 3.

Planting operation cash costs as affected by production system treatments, 2006.

View Table

Table 4.

Cash costs and returns for organic production of ‘Duke’ and ‘Liberty’ blueberry on raised beds.^z

View Table

Table 5.

Cash costs and returns for organic production of ‘Duke’ and ‘Liberty’ blueberry on flat beds.^z

View Table

Cultivar.

‘Liberty’ is a patented cultivar and cost $0.30 more per plant or $1315 more per hectare than ‘Duke’ in 2006 (Table 3). ‘Duke’ and ‘Liberty’ are popular cultivars for early and mid- to late-season fresh market production, respectively (Strik and Finn, 2008). ‘Liberty’ required an average of 24% to 25% more time or labor cost to prune in 2008 to 2009 (Table 6) but on average produced 5% higher return than ‘Duke’ over the 2 years (Table 2), thus increasing net cumulative returns by $16,441 and $16,376/ha for raised and flat beds, respectively (Tables 4 and 5). The fruit from both cultivars sold for $3.20/kg in 2008 and $2.76/kg in 2009. The was a significant interaction between cultivar and fertilizer applied in both years, because the higher rate of fish emulsion reduced gross returns considerably in ‘Duke’ but not in ‘Liberty’ (Table 2).

Table 6.

Operation cash costs for organically grown blueberry established on raised and flat beds, Years 1 to 3 (2007 to 2009).

View Table

Fertilizer.

The highest gross returns in each bed type, mulch, and cultivar were produced with either a low rate of fish emulsion or a high rate of feather meal (Table 2). Contrasts indicated that the source of fertilizer (feather meal produced higher returns than fish emulsion) was more important than the rate of fertilizer. Fish emulsion cost more than feather meal and was much more expensive to apply considering seven applications were needed compared with two in the feather meal treatment (Table 7). Also, fish emulsion had to be applied with specialized equipment that would not corrode. To reduce application costs, fish emulsion could be applied more cost-effectively by fertigation through a drip irrigation system. In our study, the feather meal fertilizer treatment required additional N, applied using fish emulsion, in 2007, which increased costs by $573/ha. Additional fish emulsion was not required in subsequent years when the first application of feather meal was done earlier in the season.

Fertilizer source or rate had no significant effect on weed presence or management costs (data not shown). In lowbush blueberry, an overhead-irrigated, non-mulched production system, higher rates of N fertilization increased weed presence (Smagula et al., 2009). In our study, the fertilizers were applied on top of the organic mulches and the sawdust around the plant (“planting hole”) in the weed mat plots during the establishment years of this study. In a commercial planting, fish emulsion may be applied through the drip irrigation system. It is unlikely that fertilizer placement had an impact on weed presence (Broschat, 2007).

Mulch and weed management.

We did not use mechanical methods for in-row weed control in this study. Sciarappa et al. (2008) were able to use a hand-operated rotary cultivator in the row, between blueberry plants, successfully for weed control on flat ground plantings in New Jersey. However, the superior performance of blueberry plants on raised beds in this study, likely as a result of improved soil aeration and drainage (Larco, 2010), led to greater returns on raised beds (Tables 2 and 4). Weed control on raised beds is limited to hand-pulling, mulching, or the use of herbicides to maintain integrity of the bed. Compost + sawdust mulch and weed mat produced higher yields and gross returns than sawdust only in both years (Table 2). In Georgia, rabbiteye blueberry plants (V. virgatum Ait.) established with organic mulches had a similar yield to those with weed mat in the first 2 years of establishment but greater yield in Years 3 to 5 (Krewer et al., 2009). In our study, determining the longer-term effects of mulch on gross and net returns will be an important aspect of economic sustainability.

Establishment (Table 3) and weed management (Table 8) costs differed among mulch types. Total installation costs (product and labor) for the weed mat treatment were $2138/ha for weed mat (with product cost amortized over 5 years) plus $927/ha for placing sawdust in the hole left for planting. The product and installation costs for sawdust and compost + sawdust mulches were $4497 and $5652/ha for the raised bed plantings, respectively. Planting on flat ground required less organic mulch, which reduced product cost (Table 3). Planting into organic mulches was less expensive ($0.40 per plant, labor) than planting through the holes cut into weed mat ($0.75 per plant) (Table 3).

Table 7.

Fertilizer treatment cash costs for organically grown blueberry as affected by mulch type, Years 1 to 3 (2007 to 2009).

View Table

Table 8.

Weed management cash costs for organically grown blueberry as affected by mulch type, Years 1 to 3 (2007 to 2009).

View Table

Weed management varied with the type of mulch used. Only hand-pulling of weeds was used in the sawdust and weed mat treatments, whereas various organically approved methods or products in addition to hand weeding, when necessary, were used in the compost + sawdust treatment. Overall, labor costs were lowest with weed mat, because weeds only emerged in the “planting hole” area around the crown of the plant. Weed control costs for the weed mat were higher in 2008 than in 2007 but lower in 2009 as the plants aged and filled in the hole cut in the weed mat for planting (Table 8). In contrast, labor costs for weed control increased with planting age with sawdust mulch to $953/ha in 2009.

Weeds were most prevalent in the compost + sawdust treatment (Larco, 2010), increasing weed management costs (Fig. 4). Others have documented better emergence and growth of weed seeds in compost as compared with sawdust or wood chip mulches (Burkhard et al., 2009). In our study, the integrity of the sawdust layer on top of the compost was not well maintained as a result of hand-pulling of weeds bringing compost to the surface and wind erosion of the sawdust layer, especially on raised beds. Weed seeds likely germinated more readily in the moist, nutrient-rich compost layer.

Symptoms of burn down of broadleaf weeds and edges of grass cover crop in an organic blueberry production systems planting in the first growing season, 20 June 2007. ‘Duke’ grown with compost + sawdust mulch and fertilized with the high rate of feather meal.

Citation: HortScience horts 47, 7; 10.21273/HORTSCI.47.7.866

• Download Figure

View Full Size

Symptoms of burn down of broadleaf weeds and edges of grass cover crop in an organic blueberry production systems planting in the first growing season, 20 June 2007. ‘Duke’ grown with compost + sawdust mulch and fertilized with the high rate of feather meal.

Citation: HortScience horts 47, 7; 10.21273/HORTSCI.47.7.866

• Download Figure

Symptoms of burn down of broadleaf weeds and edges of grass cover crop in an organic blueberry production systems planting in the first growing season, 20 June 2007. ‘Duke’ grown with compost + sawdust mulch and fertilized with the high rate of feather meal.

Citation: HortScience horts 47, 7; 10.21273/HORTSCI.47.7.866

• Download Figure

The methods of weed control varied in their effectiveness. Propane flaming was effective on very small weeds but could not be used in the dry summer months in this drip-irrigated field as a result of risk of setting the mulch on fire. We stopped using this control method after 2008. Using a concentrated solution of acetic acid (vinegar) was effective against small weeds, especially broadleaf species, when application was followed by at least 1 or 2 d of hot, dry weather. Although results seemed to indicate vinegar application effectively controlled larger weeds initially, these often regrew (Figs. 4 and 5). Although the product cost was reasonable for this method of control (e.g., $249/ha), special equipment had to be constructed and used to prevent corrosion of parts, which added considerable cost and labor for operation (Table 8; Fig. 5). In 2009, small weeds were effectively controlled with post-emergence applications of lemon grass oil, but frequent applications to catch emerging weeds were required and hand-weeding was still needed to keep the plots free of weeds by the end of the growing season. Our experiences in this study, on the advantages and disadvantages of using organic and inert mulches, flaming, and organic herbicides, agree with the summary provided by Granatstein and Mullinix (2008).

System used to apply concentrated acetic acid for spot control of weeds in the row, 2008.

Citation: HortScience horts 47, 7; 10.21273/HORTSCI.47.7.866

• Download Figure

View Full Size

System used to apply concentrated acetic acid for spot control of weeds in the row, 2008.

Citation: HortScience horts 47, 7; 10.21273/HORTSCI.47.7.866

• Download Figure

System used to apply concentrated acetic acid for spot control of weeds in the row, 2008.

Citation: HortScience horts 47, 7; 10.21273/HORTSCI.47.7.866

• Download Figure

In Year 3 of the study, when weed pressure was greatest, total weed control costs were five- to 12-fold greater in the sawdust mulch (hand-pulling only) and compost + sawdust mulch (various methods of weed control) as compared with the weed mat, respectively (Table 8). Although cumulative costs for establishing cherries were higher when using weed mat than for bare ground production systems, this was quickly offset by higher returns (Yin et al., 2007). Here we found that the much lower weed management costs in addition to amortizing the cost of weed mat over its expected 5-year life offset any added establishment cost by the end of the first growing season when compared with the industry standard sawdust mulch treatment (Tables 4 and 5).

Other maintenance costs.

Pruning costs varied with mulch after the first year after planting (P < 0.0001) but were unaffected by mulch immediately after planting or following the second year after planting (Table 6).

The irrigation system was installed in Year 1 and irrigation labor and maintenance were similar among treatments. Higher soil water content was measured on flat ground than on raised bed plantings (Larco, 2010), as has been observed by others (Spiers, 1995); raised beds thus required more irrigation to maintain a soil water content within a suitable range for blueberry (Larco, 2010). The cost for pumping the additional irrigation water required on raised beds relative to flat beds was relatively small (Table 6).

Weed mat plots required approximately twice as much irrigation to maintain the same soil water content as plots mulched with compost + sawdust with additional pumping costs of $69/ha on raised beds (Table 6). Soil water content was less under weed mat than under wood chip mulch in apples [Malus ×sylvestris (L.) Miss. var. domestica (Borkh.) Mansf.; Choi et al., 2011] or compared with bare soil in cherries (Prunus avium L.; Yin et al., 2007), although the impact on irrigation requirement was not addressed in their studies.

A trellis was installed at the beginning of the second year after planting (Fig. 2), as is typical for commercial fields to improve hand- and machine-harvest efficiency (Strik and Buller, 2005). The trellis also reduced wind damage in ‘Liberty’ (B.C. Strik, personal observation). The labor to install the trellis in the second growing season was $1853/ha.

The grass cover crop between the rows was mowed 11 times per growing season, and edges were maintained with a string trimmer, hand-hoeing, or targeted applications of vinegar (Fig. 4). Hand-hoeing to prevent encroachment of the grass cover crop into the planting row was done in 2007 and 2008 at a cost of $519 to 667/ha. In all years, vinegar applied to the edges of the cover crop was successfully used but was expensive relative to using a power weed trimmer (Table 6).

Additional production costs included standard replanting costs (estimated 2% blueberry plant loss in 2008), organic certification and inspection costs (Oregon Tilth, Corvallis, OR), annual soil and tissue analysis, equivalent land rent fees, and rental of bee hives and bird control measures during the fruit production years (2008 to 2009; Table 6).

Production costs and returns.

The cash costs and returns for ‘Duke’ and ‘Liberty’ grown with the three mulch treatments and fertilized with the best performing treatments (29 kg·ha⁻¹ of N as fish emulsion and 57 kg·ha⁻¹ of N as feather meal; Table 2) on raised or flat beds are shown in Tables 4 and 5, respectively. Total establishment costs for the first two growing seasons when there was no fruit harvest are presented. Harvest costs in 2008 and 2009 included picking labor, loading and hauling, picking buckets, and use of a bin trailer. Gross returns reflect fruit sales from harvested yield.

Cumulative net returns ranged from –$32,967 (‘Duke’ grown on weed mat, fertilized with feather meal) to –$50,352/ha (‘Duke’ grown with compost + sawdust, fertilized with fish emulsion) when grown on raised beds (Table 4). On flat plantings, cumulative net returns ranged from –$34,320 (‘Duke’ grown on weed mat, fertilized with feather meal) to –$52,848/ha (‘Liberty’ grown with compost + sawdust, fertilized with fish emulsion) (Table 5). It is typical in blueberry production to have negative cumulative net returns at this stage of planting establishment (Julian et al., 2011a, 2011b).

Plantings established on raised beds had a greater cumulative net return than those grown on flat beds. Although ‘Duke’ had lower establishment and pruning costs, ‘Liberty’ produced higher yield, which led to higher net returns in all treatment combinations except feather meal with either weed mat or compost + sawdust mulch (Table 5). When using raised beds, the highest net cumulative returns in this study were found in plots mulched with weed mat and the lowest in those mulched with compost + sawdust, attributable mainly to higher production costs for weed management.