Ohio State University Extension Bulletin

Fruit Crops: A Summary of Research 1998

Research Circular 299-99


Yield, Berry Quality, and Economics of Mechanical Berry Harvest in Ohio

Richard C. Funt, Thomas E. Wall and Joseph C. Scheerens

Introduction

Mechanical harvest systems for blueberries and brambles (raspberries and blackberries) have gained wide acceptance in recent years due to the increased demand for both fresh market and processed berries, the increased acreage planted to berry crops, the scarcity of seasonal labor, and the need to reduce production costs in a competitive world market (8). Early mechanical harvest systems, widely used by the late 1950s, employed mechanical hand-held vibrators powered by compressed air or batteries to remove fruit which was then retrieved in a cloth-covered catching frame positioned at the base of the plant. Such simple systems increased harvest efficiency (time) by more than 250% and reduced harvest cost by 55% (21). Since then, advances in harvester design and technology have enhanced overall harvester performance, improved the quality of mechanically harvested fruit, and reduced the incidence of mechanical damage to fruiting plants (8). The development of cultivars and production systems optimized for mechanical harvest have also contributed to the popularity of the harvester among growers. Currently, mechanical harvesters are being sold in countries throughout the world, including the United States of America, Canada, Scotland, Russia, Chile, Argentina, and Poland (8).

Technological advances notwithstanding, the processing industry remains the primary outlet for most mechanically harvested berries. For this reason, mechanical harvest of blueberries and bramble fruit is perhaps most prevalent in the Pacific Northwest, where weather conditions are relatively dry during harvest and the processing industry infrastructure is well-developed. Oregon growers, for instance, now mechanically harvest up to 98% of their red raspberries (7). Although mechanical berry harvest has become a standard practice in the Pacific Northwest, a minor, but significant portion of the total berry crop is still hand harvested when weather conditions are favorable and then sold at premium returns through the fresh market. This ability to alternate harvesting techniques in response to markets and weather conditions has led to the continued viability of the industry.

Historically, Ohio's small-fruit industry has been based on the production of high-quality fruit for the fresh market. Although small-fruit acreage within the state is currently limited (9, 18), recent publicity regarding the health benefits of berries and the proximity of Ohio to large metropolitan areas on the East Coast suggest that Ohio's industry is positioned for expansion. During a recent strategic planning exercise, members of the Ohio Fruit Growers Society and their collaborators identified four interrelated primary components necessary for the growth and long-term stability of the small-fruit industry in Ohio:

Undoubtedly, the design or adaptation of a mechanical harvester for Ohio will be a key element in the successful development of dual production, harvesting, and marketing systems in this region. Harvesters developed for use in the Pacific Northwest may be suitable for the Ohio industry, but data regarding their efficiency or that of other harvesters under Ohio conditions is almost nonexistent (i.e., prior to this study in 1995, blueberries had been mechanically harvested by Ohio growers on a limited basis, but brambles had not). Moreover, unlike the Pacific Northwest, wet weather conditions often prevail during the Ohio harvest season. Heavy rainfall can result in softer, more easily damaged fruit, can increase Botrytis cinerea (grey mold) infection rates and severity, and can reduce the number of operating days (8). Wet weather can also increase the susceptibility of bramble primocanes to mechanical damage and hasten their subsequent cane death by Leptosphaeria coniothyrium. Ultimately, consideration of Ohio's unique climatic conditions will most likely affect the evaluation of the mechanical efficiency and economic viability of a given harvester design.

The major objective of this study was to evaluate, under Ohio conditions, the suitability of a popular mechanical harvester designed for the Pacific Northwest (19). Target crops included black raspberries, highbush blueberries, fall-bearing red raspberries, and thornless blackberries. Yield, performance parameters, fruit quality, and costs associated with the acquisition, maintenance, and operation (harvest expenses) of the harvester were explored.

Materials and Methods

Mechanical Harvesters


Frontal view of Littau (Model FR9508) Harvester drum details of Littau (Model FR9508) Harvester
grading belt details of Littau (Model FR9508) Harvester transport of Littau (Model FR9508) Harvester
Figure 1. The Littau (Model FR9508) Harvester at Trial 3 ('Chester' thornless blackberries), Piketon, Ohio. Frontal view (top left); drum details (top right); grading belt details (bottom left); harvester transport (bottom right).

A Littau (Model FR9508) self-propelled, over-the-row, "shaking drum with fingers"-type harvester (Figure 1) was leased from the Littau Berry Harvester Co. (Salem, Oreg.) through the support of growers, the Ohio Department of Agriculture, and The Ohio State University (19). In the front of the machine, vertically oriented spiked-drum shakers oscillated in a horizontal plane. As the machine moved down the row, these drums rotated freely and the oscillation caused ripe fruit to drop onto a catching surface that opened and closed around the canes. Berries rolled into cups and were conveyed vertically to a cleaning and sorting belt. Sorted fruit was collected in field containers, which were removed from the machine by hand at the end of the row.

In addition to the Littau machine, a grower-owned BEI (Model 'Little Blue,' Blueberry Equipment Inc., South Haven, Mich.) tractor-mounted, "slapper"-type mechanical harvester was used to harvest blueberries.

Trial Sites and Experimental Design

Trial sites, cultivars harvested, dates of harvest, host growers, planting descriptions, and trial experimental designs are listed in Table 1. Experiments were performed using mature plantings at all sites. Trials 1, 3, 4, and 5 were conducted concurrently with the first harvest of their respective sites during the 1995 season, whereas experimental plots used in Trial 2 had been harvested by hand three times prior to experimentation. In Trials 1 and 5, fruit removal efficiency during harvest (based on total, marketable, and cull fruit yields) was evaluated using eight uniform small-scale plots arranged in randomized complete block designs (RCBs), with two harvesting treatments (hand and Littau Harvester) and four replications per treatment. Two similar RCB designs (Tests A and B) were used for Trial 4 (highbush blueberries); Tests A and B contrasted the efficiency of fruit removal by hand picking with that of mechanical harvest using the Littau and BEI harvesters, respectively. Estimates of fruit harvest rates for the Littau harvester were determined in Trials 1, 2, and 4 using large-scale plots (60-452 ft. of row, plots not replicated). At each site, machine speed (mph) was ascertained, and fruit yields and times to harvest were recorded. Plot descriptions (i.e., plot size or composition) for each trial are listed in Table 1. No experimental parameters were measured during the harvest of Trial 3 (blackberries), as it was intended only as a field demonstration of the Littau harvester for growers.

Table 1. Blueberry and Bramble Mechanical Harvest Trial Dates, Hosts, Locations, and Field Descriptions, Trial Designs, and Plot Descriptions, 1995.

  Trial Design andPlot Description
Trial Crop, Date, Host, Location, and Field Description Fruit Removal Efficiency Machine Harvest Ratez
1y Crop: 'Bristol' black raspberries
Date: June 30
Host: Dale Stokes
Location: Stokes Berry Farm, Wilmington, Ohio
Plant Spacing: 3' X 12'x
Trellising: none
Design: RCB
Treatments: Littau harvester, hand harvest
Reps: 4
Plot size: 20 ft of row
Design: none
Treatments: Littau harvester
Reps: 1
Plot size: 60 ft of row
2y Crop: 'Bristol' black raspberries
Date: July 11
Host: Dale Stokes
Location: Stokes Berry Farm, Wilmington, Ohio
Plant Spacing: 3' X 12'
Trellising: none
n.d.w. Design: none
Treatments: Littau harvester
Reps: 1
Plot size: 250 ft of row
3 Crop: 'Chester' thornless blackberries
Date: Aug. 17
Host: The Ohio State University
Location: Piketon Research and Extension Center, Ohio
Plant Spacing: 8' X 12'
Trellising: 2 wire I-trellis
n.d. n.d.
4 Crop: 'Elliot' highbush blueberries
Date: Aug. 23
Host: Steven Bielstein
Location: The Blueberry Patch, Mansfiel, Ohio
Plant Spacing: 4' X 12'
Trellising: none
Design: RCB
Treatments
Test A: Littau harvester, hand harvest
Test B: BEI harvester, hand harvest
Reps: 4
Plot size: 2 mature plants or 10 ft of row
Design: none
Treatments: Littau harvester
Reps: 1
Plot size: 452 ft of row
5 Crop: 'Heritage' red raspberries
Date: Sept. 9
Host: Robert Rothchild
Location: Rothchild Berry Farm, Urbana, Ohio
Plant Spacing: not applicable (hedgerow-planted)
Trellising: none
Design: RCB
Treatments: Littau harvester, hand harvest
Reps: 4
Plot size: 20 ft of row
n.d.
z Littau harvester (Model FR9508)
y Trial 1 and Trial 2 harvests of 'Bristol' black raspberries were obtained from different experimental sites on the samefarm; harvest dates correspond with the first and fourth harvest (i.e., beginning and end) of the season, respectively.
x Spacings indicated correspond to within row x between row measurements.
w n.d. = not determined. Because Trial 3 was intended as a field demonstration of the Littau harvester only, no harvest parameters were measured.

Harvesting Procedures

Members of the Ohio Small Fruit Research and Extension Team hand-harvested designated plots at each site prior to machine harvest. Ripe fruit (judged by appearance) were carefully removed from all plants within each plot, collected in containers appropriate for each crop, then placed in the shade. Machine harvest of the small- and/or large-scale plots at each site was accomplished using harvester settings that were determined during preliminary harvester runs. Machine-harvested fruit were collected in commercial lugs without hand sorting on the harvester's grading belt. Lugs were immediately placed in the shade.

Soon after harvest, berries from all harvest treatments were sorted and weighed; total weight, weight of marketable berries (whole rounded berries without grey mold, defects), culls (unripe or damaged berries, berries with stems attached), and trash (stems, leaves, and other debris, blueberries only) were recorded. Yield data were obtained in the field. In addition, blueberries (Trial 4) were sorted by machine in a standard packing shed.

Fruit-Quality Evaluations

Quality evaluations were obtained on a per-plot basis for fruit that was hand or machine harvested from small-scale plots in Trials 1 and 5 and on a per-treatment basis for those harvested in Trial 4. Black raspberry fruit were sampled directly from picking containers, whereas blueberry and red raspberry fruit were mechanically or hand sorted respectively to remove culls prior to sampling for quality analyses. Red raspberry fruit firmness (Trial 5) was evaluated on site. All other quality analyses were performed on fruit samples (500 g - 1 kg) that were randomly chosen from the total harvest of each treatment (blueberries) or plot (black raspberries and red raspberries), placed in plastic bags, and then transported in ice chests to the Crop Quality Evaluation Laboratory on the Wooster, Ohio, campus of The Ohio State University's Ohio Agricultural Research and Development Center. Laboratory measurements of each quality parameter were obtained using appropriately sized, randomly chosen subsamples.

In the laboratory, the percentage of immature black raspberry fruit (Trial 1) was determined gravimetrically; fruit with torus and pedicel attached were considered to be immature. Aliquots of the macerate from two 20-berry subsamples per plot were used to measure soluble solid levels by refractometry (Abbe, Mark II, Model 10480s/n, Keene, N.H.). Titratable acidity levels were determined on 4 g aliquots of the macerate from two 20-berry subsamples per plot using standard techniques described by Perkins-Veazie and Collins (20). Color reflectance parameters were measured on 180 ml subsamples per plot (three readings per subsample) using a Hunter Color Difference meter (Model 25, Reston, Va.). Raw reflectance values of L*, a*, and b* were transformed according to Setser (25).

Differences in blueberry fruit firmness (Trial 4) were detected using four five-berry subsamples per treatment. Individual berry firmness (the force required to penetrate the epidermis) was determined on an Instron (Model 1101, Canton, MA) firmness tester equipped with a 500 g transducer and a 3.2 mm diameter probe traveling at 50 mm/min. Blueberry soluble solid contents, titratable acidity levels, and color reflectance values were measured using techniques similar to those described earlier. Storage performance of hand- and machine-harvested blueberries (four storage trials per treatment) were evaluated by transferring a weighed quantity ("H80 g/storage trial) of intact berries to new fiberboard pint containers, then storing them in air at 3ºC for five days. Containers were uncovered, but protected against air currents during storage. Percentage moisture loss was calculated from sample weights before and after storage; percentage fruit loss was determined as the ratio of the weight of damaged, shriveled, and moldy berries to after-storage weight.

Red raspberry fruit firmness (Trial 5, measured on site) was evaluated on five-berry subsamples per plot using an Ametec AccuForce II Gauge (Model ML4432-5, Largo, Fla.) equipped with a flat-surfaced probe. Individual berry firmness (the force required to close its central cavity opening) was measured with this gauge following the technique described by Perkins-Veazie and Collins (20). Soluble solids, titratable acidity, and color reflectance assessments were obtained in the laboratory using techniques identical to those described previously.

Economic Evaluations

In order to compare the economic efficiency of hand and machine harvest, a detailed economic analysis was prepared as follows:

Fixed and Variable Machine Costs
Fixed and variable machine costs were calculated (Table 2) using methodology and rationale reported previously (10). Acquisition, maintenance, and operating costs for both the Littau and BEI harvesters were based upon a projected seven-year life span. Interest, depreciation, repair, insurance, housing, fuel, and operating labor costs and salvage values were determined in an identical manner for each machine.

Table 2. Fixed and Variable Costs for Acquisition, Maintenance, and Operation of Mechanical Berry Harvesters.

  BEI Harvester Systemz
Costs Littau Harvestery Harvester Tractor
Acquisition Costs
   Machinery ($)
78,000
15,000
20,000
   Depreciation ($/yr)x
10,029
1,928
2,571
   Intrest/yr ($/yr)w
4,290
825
1,100
Maintenance Costs
   Repairs ($/yr)v
2,786
535
714
   Insurance and housing ($/yr)u
48
36
36
Acquisition and Maintenance Cost
   Total ($/yr)
17,153
3,324
4,421
Operation Costs
   Fuel/oil/lube ($/hr)t
1.32
2.26
   Labor ($/hr)s
36.74
36.74
   Operating Cost Total ($/hr)
38.06
39.00
z BEI harvester (Model 'Little Blue'); pull-type, propelled by a 38-horsepower diesel tractor.
y Littau harvester (Model FR9508); self-propelled, equipped with a 22-horsepower diesel tractor.
x Based on straight-line depreciation over seven years with 10% salvage value.
w Based on intrest at 10% per year on the average balance for seven years.
v Based on 25% of initial machine cost.
u Housing costs estimated at $0.20/ft2/yr.
t Fuel costs estimated at $1.10/gal and 0.045 gal/horsepower/hr of operation; oil and lubrication costs based on 20% of fuel cost.
s Based on salaries of one operator (driver) at $14.30/hr and three fruit handlers at $7.48/hr.

Mechanical Harvest Costs
Expected machine harvest rates (lbs. of fruit/hr.) were determined from optimal harvester speeds and subsequent yield data from large-scale plots harvested during the 1995 field demonstrations. Machine-harvest rate calculations assumed harvester speeds of 0.57 mph and 0.25 mph for the Littau and BEI machines, respectively. These values were based on actual recorded maximum speeds in the field. A six-min/hr (10%) turn around and down time was added for expected shifts of containers and harvest-crew breaks.

Harvest rates and fixed and variable machine cost statistics were then used to calculate the following: cost per hour of operation; cost per acre at 12- or 10-ft.-row spacings; and cost per harvested pound at fixed harvest rates of 1,000, 500, and 250 pounds of fruit per hour for the Littau harvester and 500, 250, and 125 pounds of fruit per hour for the BEI harvester, targeted to represent expected rates for first and subsequent harvests of the same field, respectively (Table 3). All costs were calculated for 40, 80, or 120 hours of estimated seasonal usage (i.e., conservative estimates based upon the modest berry acreage in Ohio and an anticipated limited initial acceptance of the harvester by growers).

Table 3. Variable Costs of Mechanical Harvesters Operation Per Hour, Per Pound, and Per Acre of Harvested Fruit at Three Levels of Seasonal Harvester Use.

  Seasonal Harvest Use (hrs/yr)
Harvester and Cost Variable 40 80 120
Littau Harvesterz
   Cost/hr of operation ($)y
466.89
252.47
181.00
   Cost/acre harvested ($)x
      12 ft row spacings
625.86
338.43
242.63
      10 ft row spacings
750.63
405.90
291.00
   Cost/lb fruit harvested ($)w
      250 lbs/hr
1.81
1.01
0.72
      500 lbs/hr
0.93
0.50
0.36
      1,000 lbs/hr
0.47
0.25
0.18
BEI Harvesterv
   Cost/hr of operation ($)u
177.36
117.39
94.33
   Cost/acre harvested ($)t
      12 ft row spacings
542.39
359.00
288.47
      10 ft row spacings
649.67
430.00
345.53
   Cost/lb fruit harvested ($)s
      125 lbs/hr
1.41
0.93
0.75
      250 lbs/hr
0.71
0.47
0.38
      500 lbs/hr
0.35
0.23
0.19
z Littau harvester (Model FR9508); self-propelled, equipped with a 22-horsepower diesel tractor.
y Estimates based on fixed and variable costs for the Littau harvester (Table 2). Estimates do not include costs for machinery or personnel transport from farm to farm.
x Estimates calculated from cost per hour estimates and machine-harvest rates of 0.746 and 0.622 acres per hour for 12 ft. and 10 ft. between row spacings, respectively; machine-harvest rates based upon a machine speed of 2,709 ft/hr [0.57 mph (or 3,010 ft/hr) - 10% turn around and down time] and 3,630 and 4,356 linear ft. of row/acre for 12 ft. and 10 ft. between row spacings, respectively.
w Estimates caculated from cost per hour of operation for the Littau harvester.
v BEI harvester (Model 'Little Blue'); pull-type, propelled by a 38-horsepower diesel tractor.
u Estimatesbased on fixed and variable costs for thee BEI harvester (Table 2). Acquisition and maintenance costs for the tractor were reduced to reflect a presumed additional useof 40 hours per year on cultural tasks not related to harvest(i.e., tractor cost per hour of operation calculated at 80, 120, and 160 hours use); harvest estimates do not include costs for machinery or personnel transport from farm to farm.
t Estimates calculated from cost-per-hour estimates and machine-harvest rates of 0.327 and 0.273 acres per hour for 12 ft. and 10 ft. between row spacings, respectively; machine-harvest rates based upon a machine speed of 1,188 ft/hr[0.25 mph (or 1,320 ft/hr) - 10% turn around and down time] and 3,630 and 4,356 linear ft. of row/acre for 12 ft. and 10 ft. between row spacings, respectively.
s Estimates calculated from cost per hour of operation for the BEI harvester.

Current and Projected Hand-Harvest Costs
A range of hand-harvest costs per pound of fruit was estimated based on three wage rates and four picking rates for bramble and blueberry fruit (Table 4). The lowest wage rate ($5.92/hr) used reflected the current minimum wage ($5.15) plus 15% benefits; the two additional wage rates ($7.07 and $8.22) were included to compare current hand-labor costs with projected increases in wages and benefits over the next seven years. The standard picking rate (64 pts/day or 6 lbs/hr) for red raspberry and highbush blueberry was derived from grower records for an average picker's performance over the entire season. Similar records were used to calculate the standard picking rate (64 qts/day or 12 lbs/hr) for black raspberries. Additional picking rates were included to reflect poor and superior picker performance.

Table 4. Projected Labor Cost for Hand Harvested Red Raspberries, Black Raspberries, and Blueberries at Three Rates and Four Picking Ratesz.

  Picking Rate (lbs/hr)y
Wage Ratex 6 7.5 12 15
  Labor cost ($/lb of fruit harvested)
$5.92/hr
0.83
0.66
0.42
0.33
$7.07/hr
1.02
0.82
0.51
0.41
$8.22/hr
1.21
0.97
0.61
0.49
z Standard picking rates (lbs/hr) were determined using grower records for the average daily volumes harvested by an average picker over the entire season. Daily volumes were converted to hourly weights as follows: red raspberry and highbush blueberry -- 64 pts/day ÷ 8 hr/day = 8 pts/hr, 8pts/hr X 0.75 lbs/hr; for raspberru -- 64 qts/day ÷ 8 hr/day = 8 qts/hr, 8 qts/hr X 1.5 lbs/qt = 12 lbs/hr.
y For red raspberry and blueberry, hourl rates of 6 and 7.5 lbs/hr correspond to 8 and 10 pts/hr, respectively; for black raspberry, hourly rates of 6, 12, and 15 correspond to 4, 8, 10 qts/hr, respectively.
x Wage rates are based on base pay rates of $5.15/hr, $6.15/hr, or $7.15/hr plus 15% benefits, respectively

Sensitivity Analysis of Hand- vs. Mechanical-Harvest Costs
To delineate the conditions under which fruit harvest using the Littau harvester would be more economically viable than hand harvesting, a sensitivity analysis diagram was prepared from data presented in Tables 3 and 4. The diagram relates hand vs. machine costs per pound of harvested fruit using projected hand-harvest costs per pound of fruit and costs per pound of machine-harvested fruit at the three estimated levels of seasonal machine use (40, 80, and 120 hrs/yr) and three estimated harvest rates (250, 500, and 1,000 lbs/hr).

Statistical Analyses

All yield and fruit-quality data from small-scale plots were subjected to analyses of variance using the SAS statistical package (24) and procedures (i.e., PROC ANOVA, GLM). Parameter means were compared using the LSD statistic (P "d 0.05).

Results and Discussion

Harvester Performance

Because Ohio's current bramble and blueberry production acreages are relatively limited, mechanical harvesters that are not suited to the harvest of several crops will not likely be adopted by the industry. Even if the industry realizes its goals for expansion and the development of a processing industry, growers are likely to continue to base their operations on the culture of several crops rather than to rely upon the harvest of a single commodity to generate total farm income. For these reasons, it was important to evaluate the harvester under consideration (i.e., the Littau Berry Harvester, Model FR9508) using a variety of crop species.

Fortunately, the Littau harvester design did allow for the effective harvest of all crops tested. The harvester seemed to be especially suited for the harvest of 'Chester' thornless blackberries and, perhaps, provided fruit that were superior to those obtained from a typical hand harvest of this crop. Nearly all machine-harvested blackberries in Trial 3 were mature, firm, and intact, with little or no bruising, epidermal tearing, or juice loss to the container. Moreover, the Littau harvester was able to negotiate the blackberry trellis, easily maneuvering around five- and six-inch posts and the double-strand trellis wire.

Harvester Design and Operating Parameters
The maximum recorded operating speed for the Littau harvester was 0.57 mph (3,010 ft/hr) for all trials. Assuming a 10% correction for turn around and down time (mechanical adjustments, crew breaks, etc.), the overall harvester speed was considered to be 0.51 mph (2,709 ft/hr). At the corrected harvester speed, the Littau harvester would require 1.34 and 1.61 hours to harvest an acre planted at 12 ft. (3,630 linear ft. row/acre) and 10 ft. (4,356 linear ft. row/acre) between row spacings, respectively. When a large number of berries were ripe, such as during the first harvest of 'Bristol' black raspberries (Trial 1) and 'Elliott' blueberries (Trial 4), the Littau machine traveling at 0.57 mph within the row (0.51 mph, overall) was capable of harvesting more than 1,000 pounds of fruit per hour. However, when harvesting blueberries (Trial 4) at this speed, the machine reached its full capacity (the point at which its rotating cups were overflowing and berries fell to the ground). The Littau harvester would need to be refitted with larger cups in order to increase operating speed without loss of crop, especially in first-harvest situations where yields have been optimized by the planting of high-yielding cultivars or the use of superior cultural techniques.

In comparison, the BEI harvester used in Trial 4 traveled at an average rate of 0.25 mph (1,320 ft/hr). When BEI harvester speeds were corrected similarly for turn around and down time (i.e., 0.22 mph or 1,188 ft/hr), it was estimated that this machine would need 3.05 and 3.66 hours to harvest an acre planted at 12 ft. and 10 ft. between row plantings, respectively.

Obviously, the actual maximum harvester speed during the harvest of a given planting would be determined primarily by field conditions and the number of ripe berries present at the time of harvest. The relative skill of the machine operator and crew would also affect harvest rates. In our studies, the operator and crew were relatively unfamiliar with operating the Littau harvester, and thus data reported herein may represent the minimum level of its performance. An experienced operator and crew may be able to adjust operating parameters (e.g., drum rotation speed) in order to maximize yields, minimize the time required for harvest of a given field or crop, and thus improve the cost efficiency of mechanical harvest.

Small-Scale Plot Yield and Fruit-Removal Efficiency
Total yields (g/plot) from hand- or machine- harvested 'Bristol' black raspberries (Trial 1) or for 'Heritage' red raspberries (Trial 5) were not statistically different (Table 5), suggesting that the Littau harvester was efficient at removing ripe berries from raspberry canes. A similar statistical pattern was uncovered for hand-and machine-harvested 'Elliott' blueberries in Trial 4 (Table 6), although plots harvested with the BEI harvester tended to yield substantially less than their hand-harvested counterparts. Machine-harvested lots of all three crops contained significantly greater cull weights and/or cull percentages than lots harvested by hand. In Trial 1, increased cull percentages resulted predominantly from the mechanical harvest of unripe berries; the dislodging and collection of immature fruit has been reported to be a disadvantage of many cane fruit and blueberry mechanical harvest systems (12, 15, 21, 27). In contrast, increased fruit bruising associated with mechanical harvest of blueberries most likely affected the increased cull percentages reported for Trial 4.

Table 5. Comparison of Fruit Removal Efficiency During Hand and Mechanical Harvest of 'Bristol' Black Raspberries and 'Heritage' Red Raspberries, 1995.

  Marketable Yield Cull Wt.z
Trial Crops, Harvest Dates,
and Harvest Methods
Total Yield (g/plot)y (g/plot) (%) (g/plot) (%)
1
'Bristol' Black Raspberries (June 30)
  Hand
3851
3537
92 ax
314 b
8 b
  Littau harvesterw
3356
2664
79 b
692 a
21 a
5
'Heritage' Red Raspberries (Sept. 9)
  Hand
331
316
95
15 b
5 b
  Littau harvester
322
281
87
41 a
13 a
z Culls consisted of unripe or over-ripe berries.
y Plot size = 20 ft. of row; Trial 1 and 5 plotshad not been harvested previously.
x Mean separation with in trials by the LSD statistic (P "d 0.05).
w Littau harvester (Model FR9508).

Table 6. Comparison of Fruit Removal Efficiency During Hand and Mechanical Harvest of 'Elliot' Blueberries (Trial 4), Aug. 23, 1995.

  Marketable Yield Cull Wt.z Trashy
Test and Harvest Methods Total Yield (g/plot)x (g/plot) (%) (g/plot) (%) (g/plot)x
A. Hand
3,768
3,305 aw
88
463 b
12 b
n.a.v
   Littau harvesteru
3,450
2,602 b
75
744 a
22 a
104
B. Hand
4,404
3,959
90
445
10 b
n.a.
   BEI harvestert
2,679
2,061
77
536
20 a
82
zCulls consisted of unripe or over-ripe berries.
y Trash consisted of stems or other debris.
x Plot size = 20 ft. of row; Trial 1 and 5 plotshad not been harvested previously.
w Mean separation with in trials by the LSD statistic (P "d 0.05).
v n.a. = not applicable.
u Littau harvester (Model FR9508).
t BEI harvester (Model 'Little Blue')

Yield figures for the harvest of 'Heritage' red raspberries (Trial 5, Table 5) were extremely low, presumably due to poor plant performance following extremely wet spring weather. Therefore, the data reported do not reflect the true potential of mechanical harvest for this crop. Additional trials re-examining the suitability of mechanical harvesting for 'Heritage' red raspberry are planned for a later date.

Large-Scale Plot Yields and Harvest Rates
The Littau harvester, designed for the commercial harvest of berry crops, is not scaled properly for small-plot research. Therefore, to more closely resemble typical machine operating conditions, within-row yields, projected per-acre yields, and fruit-harvest rates were based on the yield (pounds/plot) and harvest time (min/plot) associated with harvester runs through large-scale plots. Within-row yields for mechanically harvested 'Bristol' black raspberry (Trials 1 and 2) and 'Elliott' blueberry (Trial 4) ranged from 0.12-0.42 lbs. per linear ft., corresponding to projected per-acre yields of 436-1,525 lbs. and 523-1,830 lbs. for fields with between row spacings of 12 ft. and 10 ft., respectively (Table 7). Although the use of large-scale plots for determination of yield and harvest rate parameters was considered to improve the accuracy of these measurements, estimates of the within-row yields of 'Bristol' black raspberry (Trial 1) based upon small plot yields (Table 5) and large plot data (Table 7) were similar (i.e., 0.37 and 0.40 lbs/ft, respectively). In contrast, within-row yield estimates for 'Elliott' blueberry (Trial 4) based on small plots (0.83 lbs/ft) were almost double those calculated from the large plot data (0.42 lbs/ft) reported in Table 7. Small-scale plots for fruit removal efficiency studies in Trial 4 were chosen in row segments that contained fully shaped, uniform blueberry plants, whereas there was considerable variation in bush volume and fruit load within plant population in the field. In cases such as this, plant to plant variation validates the necessity of using large-scale plot data for reliable yield-potential and harvest-rate estimates.

Table 7. Within-Row Yields, Projected Yields Per Acre, and Fruit Harvest Rates Associated with Mechanical Harvest of Blueberries and Brambles, 1995z.

  Projected Yield  
Trial Crops, Harvest Dates Plot Size (ft of row) Harvest Duration (min/plot) Total Yield (%) Within Row Yield (lbs/ft) 12 Ft. Between Rows (lbs/acre)y 10 Ft. Between Rows (lbs/acre)x Fruit Harvest Rate (lbs/hr)w
1'Bristol' Black Raspberries (June 30)
60
1.33
23.35
0.40
1452
1742
1084
2'Bristol' Black Raspberries (July 11)
250
5.54
29.11
0.12
436
523
325
3'Elliot' Highbush Blueberries (Aug. 23)
452
10.00
189.06
0.42
1525
1830
1138
z All data and calculations based upon performance of the Littau harvester (Model FR9508).
y Calculated from within-row yields based on a conversion factor of 3,630 linear ft. of row/acre.
x Calculated from within-row yields based on a conversion factor of 3,630 linear ft. of row/acre.
wFruit harvest rates from within-row yields based on a machine speed of 2,709 ft/hr [0.57 mph (or 3,010 ft/hr) - 10% turn around and down time].]
vTrial 1 and Trial 2 harvests of 'Bristol' black raspberries were obtained from different experimental sites on teh same farm; harvest dates correspond with the first and fourth harvest (i.e. beginning and end) of the season, respectively.

Fruit-harvest rates (lbs/hr) for the Littau harvester, based on average within-row yields and an in-row machine speed of 0.57 mph, are also presented in Table 7. Machine-harvest rates ranged from 325-1,138 lbs/hr for the Littau harvester which corresponded well with the projected range of harvest rates (250-1,000 lbs/hr) used for economic analyses (Table 3). Since machine harvest may be employed in a given field repeatedly throughout the harvest season, it was important for our study to examine the performance of the mechanical harvester at first and subsequent harvests. A comparison of yield data and fruit-harvest rate calculations for Trials 1 and 2 illustrates expected differences associated with the first and fourth (last) harvest of a 'Bristol' black raspberry planting. By the fourth harvest cycle, both within-row yield and fruit harvest rates were reduced to approximately 30% of those obtained during the initial harvest of the season. Although it may be possible to increase harvester speeds during late harvests, it is not likely that they could be successfully trebled or quadrupled (due to inefficient fruit removal and increased damage to both fruit and plants). Therefore, later mechanical harvests are likely to yield less return on invested harvest costs, and at some point during the harvest cycle, use of the mechanical harvester will not be economically feasible. Likewise, hand-harvest systems are also subject to "the law of diminishing returns." Hand-harvest rates vary among workers and with respect to berry size and yield per acre. Generally, the first berries harvested are larger than those present after several pickings. When berries are large, hand-harvest rates tend to be optimal; when berries are small and scattered, workers are unable to pick at optimal rates.

During this study, hand-harvest rates were determined only in Trial 4. Seven members of the Ohio Small Fruit Research and Extension Team were able to hand harvest eight plots of 'Elliott' blueberries yielding 9.3±1.2 lbs/plot in an average of 80.5±9.7 min/plot which corresponded to a hand harvest rate of 7.0±0.3 lbs/hr. This figure agreed well with the historically determined rate of 6 lbs/hr reported for blueberries in Table 4. When compared with actual or historically determined hand-harvest rates, harvests by machine in Trials 1, 2, and 4 were found to be 181, 54, and 163 times more efficient than harvesting by hand, respectively, on the basis of harvest time alone. Admittedly, this overwhelming machine advantage must be considered within the context of expected yields at initial and subsequent harvests, harvest costs, and expected returns for fruit harvested when evaluating the overall efficiency of either harvest system.

Fruit Quality

The continued viability of the Ohio fresh-market berry industry depends upon the production and marketing of high-quality fruit. Fruit quality at harvest will likely be of equal importance to the development of an Ohio-based processing industry as it affects, to a substantial degree, the state's potential to offer a superior product to the marketplace (9). For this reason, the quality of machine-harvested fruit was compared to that of its hand-harvested counterparts in Trials 1, 4, and 5.

Black Raspberries, Trial 1
In Trial 1, a small but significant number of immature 'Bristol' black raspberries were removed by the mechanical harvester, whereas hand-harvested fruit was uniformly mature (Table 8). Moreover, a comparison of laboratory and field data (Tables 5 and 8) suggests that approximately 90% of the machine-harvested cull berry weight per plots resulted from immature berry collection, whereas culls from hand-harvested plots were more likely to have been over-ripe or mechanically damaged (bruised or torn) during the harvesting process. As might be expected, the conspicuous presence of red, pink-tinged, and even white immature berries in machine-harvested lots had a significant effect on the color reflectance values of their respective laboratory samples (i.e., mechanically harvested samples were lighter, substantially more red/orange, and more vivid overall).

Table 8. Quality Difference in Hand- and Mechanically Harvested 'Bristol' Black Raspberry Fruit (Trial 1), June 30, 1995z.

        Color Reflectance Valuesv
Harvest Method Immature Fruity (%) Soluble Solidsx (%) Titratable Acidityw (%) L Θ Chroma
Hand
0.0 bu
8.0
0.82
10.3 b
355.7 a
1.9 b
Littau Harvestert
18.5 a
7.6
0.92
12.8 a
26.7 b
5.8 a
z Quality differences were ascertained using 500 g samples randomly chosen from the total fruit harvest of each plot. Subsamples were chosen randomly for the measurement of each parameter.
y Fruit with torus and pedicel attached were considered to be immature; values represent percentage of immature fruit by weight.
x Aliquots of the macerate from two 20-berry subsamples/plot were used to measure soluble solid levels by refractometry.
wAliquots (4 g) of the macerate from two 20-berry subsamples/plot were used to determined titratable acidity levels.
vColor reflectance parameters were measured on four 180cc subsamples/plot (3 readings/subsample) using a Hunter Color Difference meter (Model 25). Values indicate the following: L indicates lightness/darkness (0 = pure black, 100 = pure white); Θ (hue angle) indicates hue as dipicted by the degrees within a circle (0° = pure red, 90° = pure yellow, 180° = pure green, 270° = pure blue, 360° = pure red); chroma depicts relative color intensity (high values indicate vivd colors).
uMean separation with in trials by the LSD statistic (P "d 0.05).
t Littau harvester (Model FR9508).

The presence of markedly immature berries in mechanically harvested lots may have resulted from suboptimal harvester settings; that is, the machine moved too slowly down the row or the drums oscillated too vigorously, removing fruit which should have been left on the canes for subsequent harvests (B. Strik, Oregon State University, personal communication). Moreover, the appearance of immature fruit in mechanically harvested lots suggested the simultaneous harvest of "almost-ripe" fruit in even greater proportions. A high proportion of almost-ripe fruit may have resulted in low soluble solid and high titratable acidity values in the machine-harvested fruit samples reported in Table 8. Morris (15) reported that mechanically harvested lots of cane fruits usually exhibited increased soluble solid contents and reduced titratable acidity levels when compared to their hand-harvested counterparts, a trend that was reversed in this study.

Although the presence of immature berries may render machine-harvested lots unsuitable for certain processing procedures [e.g., individually quick frozen (IQF) fruit packs], their use in more thermally processed products (preserves, juice concentrates, etc.) may have minimal impact on product quality (17). Moreover, since they are highly visible, immature berries are easily removed by hand or machine sorting after harvest. Finally, the percentage of immature berries in a given harvest lot may be highly dependent on cultivar, minimized by improved cultural practices, and, most prominently, affected by the skill of the harvester operator (12, 27).

Blueberries, Trial 4
"Perhaps the most serious limitation to current mechanical harvesting technology is the damage caused to fruit" (8). Blueberries can be easily bruised during any stage of mechanical harvesting and sorting that results in impact after a vertical fall. Brown et al. (4) reported extensive bruising in ripe blueberries when the drop height to a hard surface exceeded six inches. The extent of damage is proportional to the distance the fruit fall, and bruised fruit are more subject to decay during postharvest storage (1, 3, 8, 13). In addition, the bruising process often results in a loss of fruit surface wax (bloom) and thus results in darker-appearing fruit (8).

The effects of bruising associated with mechanical harvest of blueberries was readily evident in Trial 4 (Table 9). Harvesting with either mechanical harvester resulted in berries that were significantly softer than hand-harvested fruit. Moreover, the storage performance of both mechanically harvested fruit lots was extremely poor. Moisture loss during storage was greatest in fruit harvested with the BEI machine, but Littau-harvested berries exhibited the greatest overall fruit loss. These results suggest that the mechanically harvested blueberry fruit in Trial 4 would have been suitable only for immediate processing.

Table 9. Quality Difference in Hand- and Mechanically Harvested 'Elliot' Blueberry Fruit (Trial 4), Aug. 23, 1995z.

        Color Reflectance Valuesv Storage Performanceu
Harvest Method Fruit
Firmnessy (%)
Soluble
Solidsx (%)
Titratable
Acidityw (%)
L Θ Chroma Moisture
Loss (%)
Fruit
Loss (%)
Hand
203 at
9.9 a
1.26 a
16.2 ab
269.0
1.9 b
9.5 c
56.9 c
Littau Harvestert
148 b
9.7 a
1.24 a
15.5 b
273.6
1.8 b
10.3 b
95.8 a
BEI Harvestert
152 b
9.9 a
1.16 b
16.8 a
271.7
2.3 a
11.2 a
83.7 b
z Quality differences were ascertained using 1 kg samples randomly chosen from the total fruit harvest of each treatment after mechanical sorting. Subsamples were chosen randomly for the measurement of each parameter.
y Differences in firmness were detected using four five-berry subsamples/treatment. Individual berry firmness (the force in grams required to penetrate the epidermis) was determined on an Instron (Model 1101) equipped with a 3.2 mm diameter probe.
x Aliquots of the macerate from 20-berry subsamples/treatment were used to measure soluble solid levels by refractometry.
w Aliquots (10 g) of the macerate from two 10g subsamples/treatment were used to determined titratable acidity levels.
v Color reflectance parameters were measured on 180cc subsamples/plot (four readings/subsample) using a Hunter Color Difference meter (Model 25). Values indicate the following: L indicates lightness/darkness (0 = pure black, 100 = pure white); Θ (hue angle) indicates hue as dipicted by the degrees within a circle (0° = pure red, 90° = pure yellow, 180° = pure green, 270° = pure blue, 360° = pure red); chroma depicts relative color intensity (high values indicate vivd colors).
u Storage performance was measured by storing four 80g subsamples/treatment in air at 3°C for five days. Percentage moisture loss was calculated from sampleweight before and after storage. Percentage fruit loss was determined as the ratio of the weight of damaged, shriveled, and moldy berries to that of after-storage weight.
t Mean separation with in trials by the LSD statistic (P "d 0.05).
s Littau harvester (Model FR9508).
r BEI harvester (Model 'Little Blue').

However, mechanical harvest procedures were not solely responsible for the poor storage performance of blueberries in Trial 4, as the hand-harvested controls in this storage study also exhibited relatively high levels of postharvest deterioration. Due to logistical constraints, Trial 4 was performed two to three days after much of the crop had reached optimum ripeness (i.e., harvested fruit tended to be over-ripe) which may have exacerbated fruit bruising in all treatments. In addition, because of the mechanical sorting process and a delay in transport to Wooster, fruit lots used for storage studies were perhaps exposed to ambient temperatures longer than desirable. The positive effects of rapid cooling to remove field heat and continued low-temperature storage on stored blueberry firmness and quality are well documented (1, 2, 3, 5, 6, 13, 15, 23, 26).

The 'Elliott' blueberry is generally considered to be firm and to exhibit excellent storage characteristics (22). Supporting this contention, the ratio of soluble solids to titratable acids in hand and machine-harvested fruit varied from 7.8 - 8.3 (Table 9), well within the range of the values indicated by Galletta and coworkers (11) to be optimal for good keeping quality. In previous, more stringent OARDC storage trials (i.e., storage for eight days at 3ºC followed by a shelf-life treatment at 20ºC for an additional nine days), 'Elliott' rated highest among 60 blueberry cultivars and breeding lines for storability; at the end of the treatment period, berries had lost only 14.5% moisture; 85.4% of the berries were considered to be salable; and when subjectively scored for edibility, berries were uniformly given the highest possible rating (i.e., 5 = fruit still firm, with flavor good to excellent) (J. C. S., unpublished data).

In summary, the quantitative effects of mechanical harvesting on blueberry quality in Trial 4 could not be clearly delineated due to extraneous factors such as the harvest of over-ripe berries and the prolonged postharvest exposure to ambient conditions. However, use of either the Littau or BEI harvesters presently available will most likely damage fruit to some degree, rendering the crop suitable for processing only. If Ohio growers desire the ability to mechanically harvest fresh-market blueberries, other harvester designs will have to be considered. For instance, in a prototype harvester recently developed through the cooperative efforts of USDA and BEI personnel, design advancements in both fruit removal and catchment systems (e.g., shorter drops, specially padded surfaces, etc.) have resulted in the mechanical harvest of 'Bluecrop' blueberry where 68% of the berries were of fresh-market quality (21). In contrast, conventional mechanical-harvest and the hand-harvest tests in their study yielded 22% and 77% fresh-market quality fruit, respectively.

Red Raspberries, Trial 5
Hand- and machine-harvested 'Heritage' red raspberry fruit quality was assessed using marketable berries only (i.e., cull berries had been removed prior to analysis). Consequently, few differences in quality were uncovered (Table 10). Morris (14, 15, 16) observed machine-harvested cane fruit (especially blackberry) to be more fully ripe than their hand-harvested counterparts, due primarily to the difficulty in visually distinguishing between blackberries that were ripe from those that were merely black. 'Heritage' color reflectance values (specifically L and chroma) tended to support Morris' contention, but in contrast to his experience, fruit firmness, soluble solid contents, and acidity levels were similar in hand- and machine-harvested fruit in this study (Table 10). Unfortunately, the postharvest longevity of machine-harvested red raspberries was not explored herein. However, even though harvesting techniques appeared to produce similar quality products in Trial 5, the processing market will remain the most likely outlet for mechanically harvested Ohio red raspberries (8).

Table 10. Quality Difference in Hand- and Mechanically Harvested 'Heritage' Red Raspberry Fruit (Trial 5), Sept. 9, 1995z.

        Color Reflectance Valuesv
Harvest Method Fruit Firmnessy (N) Soluble Solidsx (%) Titratable Acidityw (%) L Θ Chroma
Hand
1.14
11.3
0.74
16.2 au
12.1
15.8
Littau Harvestert
1.25
11.5
0.71
15.9 b
12.1
14.5
z Quality differences were ascertained using 500 g samples randomly chosen from the total fruit harvest of each plot. Subsamples were chosen randomly for the measurement of each parameter.
y Differences in firmness were measrured on site using five-berry subsamples/plot. Individual berry firmness (the force in Newtons required to close the central cavity opening) was determined on an Ametec AccuForce II (Model ML-4432-5) equipped with a flat-surfaced probe.
x Aliquots of the macerate from two 20-berry subsamples/plot were used to measure soluble solid levels by refractometry.
w Aliquots (10 g) of the macerate from two 20-berry subsamples/plot were used to determined titratable acidity levels.
v Color reflectance parameters were measured on four 180cc subsamples/plot (four readings/subsample) using a Hunter Color Difference meter (Model 25). Values indicate the following: L indicates lightness/darkness (0 = pure black, 100 = pure white); Θ (hue angle) indicates hue as dipicted by the degrees within a circle (0° = pure red, 90° = pure yellow, 180° = pure green, 270° = pure blue, 360° = pure red); chroma depicts relative color intensity (high values indicate vivd colors).
u Mean separation with in trials by the LSD statistic (P "d 0.05).
tLittau harvester (Model FR9508).

Economic Assessment

Most raspberries, blueberries, and blackberries grown in Ohio are currently harvested by hand for fresh-market sales, and growers are well aware of the economic advantages and disadvantages of current cultural, harvesting, and marketing systems. However, whenever new markets or new technologies are developed (such as those which might serve an emerging processing industry), growers must rely on accurate economic analyses in order to make rational decisions concerning the cultural and marketing alternatives available to them. Therefore, the authors have endeavored to present herein a detailed economic assessment of harvest costs based on field performance for the use of the Littau harvester for bramble and blueberry harvest and have included comparative data for the BEI harvester, which is suitable for mechanically harvesting blueberries.

Fixed and variable costs for the acquisition, maintenance, and operation of the mechanical berry harvesters are listed in Table 2. Projected acquisition and maintenance costs for the self-propelled Littau harvester ($17,153/year over seven years) were approximately 2.2 times higher than those calculated for the BEI machine and tractor combination ($7,747/year over seven years), whereas operational costs were nearly identical for the two machines. Admittedly, the BEI system is attractive, considering its low acquisition and maintenance costs and the advantage of employing an independent power source which can be used to perform cultural tasks other than harvesting. However, it is not nearly so versatile as the Littau machine which can effectively harvest a number of small fruit crops. As stated earlier, limited small-fruit acreages in Ohio coupled with the industry growers' tendency to base their operations on the culture of several crops make harvester versatility a consideration of paramount importance.

The cost per hour of machine operation (based directly on fixed and variable costs) varied among harvesters and among projected seasonal usages of 40, 80, and 120 hrs (Table 3). Obviously, the cost of operation diminishes greatly as seasonal usage is increased. Seasonal use of a mechanical harvester on an average berry farm in Ohio (approximate size equals two acres) would be insufficient to support the purchase and maintenance of such expensive equipment. Therefore, for economic viability, it will be important for growers to maximize machine usage through cooperative ownership or through the contracted harvest of several farms within a region. Multiple machine harvests of a given field will also need to be considered. Likewise, harvester versatility will increase the likelihood for maximum use of a single machine. In such situations, actual machine use (hours per year) is likely to exceed by two- to threefold the maximum usage level (120 hrs/yr) projected herein.

Costs per acre harvested at 12 ft. and 10 ft. between row spacings (based on costs per hour and field-determined optimal machine speeds) and costs per pound of harvested fruit were also determined at various levels of season harvester use (Table 3). Costs per pound of fruit harvested varied greatly for both machines with respect to target harvest rate (based on field data for first and subsequent harvests) and seasonal use estimates. The target harvest rates for the Littau harvester used in these calculations were twice those used for the BEI harvester based on optimal machine speeds. At a given target harvest rate, costs per pound of fruit harvested were generally lower for the BEI machine at 40 and 80 hours of seasonal use. However, apparently, this cost advantage per pound of harvested fruit is nearly lost as seasonal harvester use approaches 120 hours of operation.

Although our economic projections do not directly consider optimal machine speed as a variable, it will undoubtedly affect the overall profitability of a given mechanical harvester. For instance, even though costs per pound of harvested fruit may be similar between the two harvesters, the Littau machine will be able to more than double the total amount of blueberries harvested over a season by covering more than twice as many acres. In other words, to harvest a similar amount of blueberries, the BEI machine must be operated 2.28 times longer than the Littau harvester, thus incurring 2.28 times the operating costs.

Projected labor costs per pound of hand-harvested red raspberries, black raspberries, and blueberries are listed in Table 4. At current minimum wage/benefit rates ($5.92/hour), a pound of red raspberries or blueberries harvested by an average laborer picking at the rate of 6 lbs/hr would cost $0.83, well within the range of costs per pound calculated for the target harvest rates of the Littau harvester (i.e., $0.17/lb - $1.81/lb) and the BEI harvester ($0.15 - $1.30). Likewise, the current hand harvest cost for black raspberries ($0.42, 12 lbs/hr average picking rate) is also within the cost per pound ranges of both harvesters. Both of these relationships suggest that at some combination of harvest rate and total seasonal use, mechanical harvest costs will be below those of hand harvest.

In order to explore these relationships, a sensitivity analysis was performed; it is represented graphically in Figure 2. The shaded area of the diagram represents current and projected costs over the next several years of hand-harvest at average picker rates. Current harvest costs calculated for the Littau harvester are superimposed on hand-harvest costs for target harvest rates of 250, 500, and 1,000 lbs/hr at seasonal usage rates of 40, 80, and 120 hrs/yr. The examples presented here serve to illustrate the use of Figure 2. At low harvest rates (250 lbs/hr, representing harvester use late in the season) and low levels of seasonal usage (40 hrs/yr), machine-harvest costs per pound exceed current and projected hand-harvest costs for all three fruit crops and thus may never be profitable. Likewise, mechanical harvest at the middle harvest rate (500 lbs/hr, representing a mixture of first and subsequent harvests) and a seasonal usage of 40 hrs also exceeds current average hand-harvest costs ($0.42/lb - $0.83/lb), but may be economically viable for the harvest of red raspberries and blueberries in the future as hand-harvest costs increase over time. Mechanical-harvest costs at the middle-harvest rate and a seasonal usage of 80 hours are currently lower than those associated with the hand harvest of red raspberries and blueberries, but not black raspberries; however, the mechanical harvest of black raspberries will be economically more feasible as hand-harvest costs rise. Mechanical-harvest costs at the middle-harvest rate and a seasonal usage of 120 hours are well below hand-harvest costs for red raspberries and blueberries and slightly below those for black raspberries. Similarly, at high mechanical-harvest rates (1,000 lbs/hr, representing first harvests) and seasonal harvester usages of 80 and 120 hours, hand-harvest costs exceed those incurred by machine harvest.


Figure 2. Sensitivity analysis of hand- vs. mechanical-harvest costs 
		based on target harvest rates and hours of seasonal usage.
Figure 2. Sensitivity analysis of hand- vs. mechanical-harvest costs based on target harvest rates and hours of seasonal usage.


In short, growers who have the management skills to produce high yields and can use a machine 80 or more hours per year are likely to lower their harvest costs using a mechanical harvester. Berry quality, the development of suitable markets for mechanically harvested fruit, and rates of return per pound of fruit sold to fresh or processing markets, respectively, must also be considered before growers can make informed decisions concerning the suitability of mechanical berry harvest for their own operations.

A Mechanical-Harvester Use Scenario for Ohio Growers

Although Ohio acreages might be exclusively harvested by machine in the future, it is more likely that the industry will adopt a harvest system similar to that used in the Pacific Northwest which includes alternate use of hand labor for fresh-market harvest and mechanical harvesters to harvest berries for processing. A decision by Ohio growers to use the mechanical harvester for any given harvest will likely be based on a number of considerations, including the current-market returns for fresh or processed fruit, the harvest-ready acreage (i.e., the number of acres waiting to be harvested), the condition of the fruit in the field (i.e., degree of ripeness), weather conditions, and labor availability.

If mechanical berry harvest is to be readily adopted in Ohio, a processing-industry infrastructure will need to be developed concurrently. The nature of that industry may also affect grower enthusiasm for expanding current bramble and blueberry acreages and for adopting new production and harvesting systems. The development of cooperatively owned and managed processing facilities from which producers profit directly from the manufacture of value-added products may be far more attractive than traditional scenarios where growers sell their crops at low-profit margins to a corporate processor.

Ample evidence has been presented to indicate that the economic performance of mechanical berry harvesters depends upon relatively high overall yields, relatively concentrated fruit ripening, harvester flexibility, and extended machine use throughout the season. A machine such as the Littau harvester, which has the capability of harvesting a variety of crops, could operate over a 12-week period, beginning with black and summer-bearing red-raspberry harvest in late June and finishing in mid-September with the harvest of blueberries and fall-bearing red raspberries. Efficient use of the machine will obviously require movement from farm to farm, as in this case, the Littau harvester was placed onto a tractor-trailer and driven with relative ease more than 100 miles from trial to trial. Multiple machine harvests of the same field are also likely. Harvester speeds used in this study indicated that the Littau harvester was capable of harvesting 6.4 acres in an eight-hour day. Therefore, 25.6 acres of blueberries or 12.8 acres of brambles could be continuously harvested on a four-day and a two-day pick cycle, respectively, using a single machine.

Summary

The results of this preliminary study show that there is a potential to mechanically harvest berries in Ohio. However, further experimentation and experience with mechanical berry harvesting will be necessary before accurate economic assessments of harvester use based on the rates of return per pound of fresh market or processing fruit can be made, or the most economic scenario for the mixed hand/machine-harvest systems can be determined. Moreover, many improvements within the industry, such as controlled growth in the total number of acres planted to berry crops, market expansion, optimization of berry yields and quality through improved and standardized cultural practices, and efficient industry organization, must be realized if mechanical berry harvesting is to reach its full potential. Likewise, our knowledge of the relationships among berry production systems, berry quality, and mechanical harvesting must be expanded through additional research efforts by university personnel and cooperating growers. Admittedly, our preliminary survey of harvester capabilities was limited to the thorough testing of only one harvester design. The merits of newer mechanical harvesters and mechanical harvesting technology should also be explored, especially if the mechanical harvest of fresh-market berries becomes an industry goal.

Acknowledgments

The authors wish to acknowledge the support of the Ohio Small Fruit and Vegetable Research and Development Program, the Ohio Rural Rehabilitation Program, and the Ohio Agricultural Research and Development Center for their financial support. We are grateful to Mark and Dale Stokes for their assistance in the transport and operation of the harvester and to Meyer Equipment Leasing Company of Wilmington, Ohio. We also acknowledge Dale Stokes, Steve Bielstein, and Robert Rothchild who graciously allowed us to conduct field testing at their farms.

References

  1. Ballinger, W. E., L. J. Kushman, and D. D. Hamann. 1973. Factors affecting the firmness of highbush blueberries. J. Amer. Soc. Hort. Sci. 98:583-587.
  2. Ballinger, W. E., E. P. Maness, and W. F. McClure. 1978. Relationship of stage of ripeness and holding temperature to decay development of blueberries. J. Amer. Soc. Hort. Sci. 103:130-134.
  3. Boyette, M. D., C. M. Mainland, and W. O. Cline. 1992. Postharvest cooling and handling of blueberries in North Carolina. North Carolina State University. Coop. Ext. Svc. Bull. Ag-413-7.
  4. Brown, G. K., N. L. Schulte, E. J. Timm, R. M. Beaudry, D. L. Peterson, J. F. Hancock, and F. Takeda. 1996. Estimates of mechanization effects on fresh blueberry quality. Appl. Eng. in Agric. 12:21-26.
  5. Brun, C., J. F. Hancock, D. Kaufman, and G. C. Pavlis. 1992. Harvesting and handling of blueberries. In: Highbush blueberry production guide. M. P. Pritts and J. F. Hancock, Eds. NRAES Coop. Ext. Svc., Ithaca, New York.
  6. Ceponis, M. J. and R. A. Cappellini. 1979. Control of postharvest decays of blueberry fruits by pre-cooling, fungicide, and modified atmospheres. Plant Disease Rep. 63:1049-1053.
  7. Cross, T., B. Turner, B. Strik, and D. Kaufman. 1993. Red raspberry economics – the costs of establishing and producing red raspberries in the Willamette Valley. Oregon State University. Ext. Bull. EM 8534. pp. 1-17.
  8. Dale, A., E. J. Hanson, D. E. Yarborough, R. J. McNicol, E. J. Stang, R. Brennan, J. R. Morris, and G. B. Hergert. 1994. Mechanical harvesting of berry crops. Hortic. Rev. 6:255-383.
  9. Funt, R. C., W. D. Bash, T. E. Wall, and D. Stokes. 1997. Small fruit production strategic plan 1997-2022. The Ohio State University. Department of Horticulture and Crop Science. Unpublished.
  10. Funt, R. C., T. A. Baugher, H. W. Hogminer, and W. C. Kliener. 1992. Profitability of different apple orchard systems in the eastern United States. The Ohio State University. Ohio Agricultural Research and Development Center. Department of Horticulture Series 624.
  11. Galletta, G. J., W. E. Ballinger, R. J. Monroe, and L. J. Kushman. 1971. Relationships between fruit acid and soluble solids levels of highbush blueberry clones and fruit-keeping quality. J. Amer. Soc. Hort. Sci. 96:758-762.
  12. Martin, L. W. 1985. Cultural modifications for improving mechanical harvesting of red raspberries and blackberries. HortScience 20:1014-1015.
  13. Miller, W. R. and D. A. Smittle. 1987. Storage quality of hand- and machine-harvested rabbiteye blueberries. J. Amer. Soc. Hort. Sci. 112:487-490.
  14. Morris, J. R. 1980. Mechanically harvesting cane fruits. HortScience 15:2.
  15. Morris, J. R. 1983. Influence of mechanical harvesting on quality of small fruits and grapes. HortScience 18:412-417.
  16. Morris, J. R. 1984. Developing a mechanical harvesting system for blackberries. HortScience 19:188-190.
  17. Morris, J. R., W. A. Sistrunk, A. A. Kattan, G. S. Nelson, and D. L. Cawthon. 1979. Quality of mechanically harvested strawberries for processing. Food. Tech. 35:92-94, 96, 98.
  18. Ohio Agricultural Statistics Service. 1997. Ohio Farm Report. Vol. 97, No. 16.
  19. Ohio Fruit Growers Society. 1995. Stokes hosts raspberry harvester demonstration. Today's Grower 9(5): 3-4.
  20. Perkins-Veazie, P. M. and J. K. Collins. 1993. Quality evaluation and post-harvest procedures for small fruits. USDA. ARS. South Central Agricultural Research Laboratory, Lane, OK.
  21. Peterson, D. L. and G. K. Brown. 1996. Mechanical harvester for fresh-market-quality blueberries. Trans. Amer. Soc. Agric. Eng. 39:823-827.
  22. Penhallegon, R., N. Schulte, and J. Spiers. 1992. Plant selection and propagation. In: Highbush blueberry production guide. M. P. Pritts and J. F. Hancock, Eds. NRAES Coop. Ext. Svc., Ithaca, N.Y.
  23. Sanford, K. A., P. D. Lidster, K. B. McRae, E. D. Jackson, R. A. Lawrence, R. Stark, and R. K. Prange. 1991. Lowbush blueberry quality changes in response to mechanical damage and storage temperature. J. Amer. Soc. Hort. Sci. 116:47-51.
  24. SAS Institute. 1990. SAS/STAT user's guide. 4th Ed. Vol. 2. SAS Institute, Inc., Cary, N.C.
  25. Setser, C. S. 1984. Color: reflections and transmissions. J. Food Qual. 6:183-197.
  26. Singh, R. P. and D. R. Heldman. 1975. Factors influencing fresh-market blueberry quality. Trans. Amer. Soc. Agric. Eng. 18:156-158.
  27. Takeda, F. and D. L. Peterson. 1988. Machine harvest of eastern thornless blackberry cultivars. HortScience 23:120-123.


Back | Forward | Table of Contents