History
Citrus mechanical harvesting dates back to the 1950’s, due to the lack of harvesting labor availability; a problem that continues today. The systems that were developed early on used several unconventional methods of removing the fruit. For example, blasting water or air into the tree knocking the fruit off the stems. The historical review by decade illustrates the development of mechanical harvesters and the equipment and methods that have been designed to be used with them.
-
1950-1960
Written by Jodie D. Whitney, July 2006
Collective concern by the Florida citrus industry about the availability of labor in 1957 and harvest steadily increasing production/acreage1 for the establishment of a citrus harvesting research project at the University of Florida Lake Alfred CREC. Seasonal labor requirements were usually over 20,000 people and peaked at 35,000 people in 1980 when citrus production was 284 M boxes (13 Mt). Harvesting then, as is now, was affected by fruit type, utilization (fresh vs. processed), grove characteristics, and the cost and availability of harvest labor. Prior to the severe freezes of the 1980s, well over 50% of citrus was produced on trees 20 to 30 ft tall on the deep sand or ridge area of Central and North Florida. Most of the mechanical harvesting research was conducted in this area.
Citrus harvesting involves two operations – 1) picking or removing fruit from the tree and placing it in a field container and 2) handling the field containers in the grove and hauling the fruit out of the grove. In the 1950s, the handling and hauling operation was mechanized, reducing handling and hauling labor by at least two-thirds.1,2
After initial efforts to improve the productivity of manual harvesters, perfecting a “mechanical picker” was the focus of many projects during the 1950s, 60s, and 70s. Much of the significant work was done by a number of researchers, done in cooperation with machinery manufacturers, inventors, and chemical companies, including:
- Glenn Coppock, Florida Department of Citrus (FDOC), who initiated the project, picking aids, contact devices, limb shaker-catchframes, and air shakers;
- Pierre Jutras, UF, air shakers and limb shakers;
- Scott Hedden, USDA Project Leader, picking platforms, limb shaker-catchframes, fruit pickup machines;
- Don Lenker, USDA, limb shaker design, contact devices;
- Dale Marshall, USDA, fruit pickup machines;
- Harold Sumner, USDA, foliage and limb shakers, fruit windrow rakes and pickup machines;
- Don Churchill, USDA, fruit pickup machines;
- Jim Donhaiser, FDOC, air shakers;
- Jodie Whitney, UF, air and trunk shakers, fruit pickup machines.
- Bill Wilson, FDOC, worked on screening and testing abscission chemicals.
Initially, time and motion studies were conducted on conventional manual harvesters to determine ways to improve their productivity by replacing ladders with mobile platforms.3 Maximum productivity gains with mobile platforms were 30 to 40%, but the investment in platforms, picking aids, man positioners, etc. was never proven to be economically feasible.4
Beginning in the early 1960s, mechanical methods in the form of contact devices (those which contact the fruit to separate it from the tree) were investigated. Many of the contact devices were designed to duplicate the fruit separation techniques of manual harvesters and be suitable for fresh or processed citrus harvesting1,5,6, but they showed little potential for acceptance. The performance of contact devices was usually limited to less than 70% fruit removal because large tree canopies and limb structures limited canopy penetration.
In the early 1960s, investigations began on mass removal (shaker) harvest systems which appeared to have potential to harvest citrus acceptable for processing at a rate commensurate with cost.1 Research was directed mainly toward harvesting processed oranges which required the bulk of the harvesting labor. Development of limb, air, trunk, and foliage (canopy) shakers, with and without catchframes, began without the fruit loosening aid of abscission chemicals.7,8,9, 10,5, 11, 12,13,14,15,16,17,18 Orange removals ranged up to 90% and were generally higher with limb shakers. Yield effects on early and midseason oranges were minimal, but yield reductions on Valencia oranges (constituted about ½ Florida’s orange production and mature fruit is harvested when young fruit or next year’s crop is on the tree), shaken late in the harvest season were up to 40%.
This combination of fruit removals and particularly large yield reductions resulted in low harvesting efficiencies.1 Harvesting efficiency was defined as the percentage of crop that could be recovered mechanically compared to manual harvesting. Thus, if a mechanical harvester removed 90% of the crop and reduced subsequent yields by 10%, the harvesting efficiency would be 0.9 x 0.9 = 0.81 or 81%.
Development work on limb shakers for Valencia orange harvesting demonstrated that long-stroke (6 to 8 in), low frequency (4 to 5 Hz) shaking maximized mature fruit removal and minimized subsequent yield reductions.19 Vertical foliage shakers were developed because it was felt that, technically, they could achieve high mature Valencia fruit removals with insignificant young fruit damage (removal) by controlled shaking in the main fruit bearing zone of the tree. Field trials over the Valencia harvest season demonstrated mature fruit removals of 80 to 90% with subsequent yield reductions of 15 to 20% or an average harvesting efficiency of 75%.12 Further research on biophysical properties of citrus related to mechanical harvesting provided a better understanding of why higher shaker fruit removals were difficult to achieve and why Valencia orange yield reductions associated with shakers were more pronounced later in the Valencia harvest season.20,21
-
1970-1990
It was recognized that some mature fruit loosening would be beneficial to shakers in the fruit removal process and abscission chemical development was initiated.22 Several abscission chemicals were developed for field testing.23,1 In 1973, Acti-Aid (cycloheximide) was cleared for use on processed oranges by the EPA (Environmental Protection Agency), but could only be used on early and midseason oranges because it was phytotoxic to young Valencia fruit and flush. Another abscission chemical, CMNP (5-chloro-3-methyl-4-nitro-1H-pyrazole), known by its trade name of Release, was given an EUP (Experimental Use Permit) by the EPA and could be used on all processed oranges, but was ultimately not approved for use. In a Valencia field trial with limb shakers, harvesting efficiencies were ~ 66% without Release and 80 to 85% with Release.1,24,19
Because fruit loosening by the abscission chemicals usually resulted in considerable preharvest fruit drop 3 to 6 days post spray, machines were developed to pickup fruit from the ground.17 In the 1970s, most of the mechanical harvesting systems utilized abscission chemicals with limb, air, or trunk shakers to remove oranges to the ground. Pickup machines were designed to pickup fruit off the ground.25 Limb shaker-catchframe systems were used on old, tall, highly skirted seedling trees where visibility was good for the machine operators. The catchframe also minimized fruit damage to oranges falling from trees 20 to 30 ft tall.1
In 1974-75, the volume of mechanically harvested oranges (mostly early and midseason) peaked at 220,000 boxes (~10,000 t). Numerous problems were encountered with these systems. Cool weather (less than 70 deg F) and/or rain within 2 hours after application usually rendered chemicals ineffective. Because of nonuniform and inconsistent fruit loosening by the abscission chemicals (Acti-Aid and Release), and losses in fruit collection, harvesting efficiencies were frequently less than 95%. Chemical plus application costs were significant and ranged from $0.10 to $0.25 per90 lb box. High rates of Acti-Aid caused excessive defoliation, and oranges sprayed with either chemical had to be processed in 7 to 10 days post application to prevent spoilage. Tree damage concerns of the grower from the chemical and/or shaker were defoliation, small and large limb breakage, and bark damage.1 Harvesting system productivities were usually low. Under optimum conditions, 1 operator on the mechanical harvesting system could harvest the equivalent of 5 manual harvesters. None of the mechanical harvesting systems proved to be economically feasible.
In the late 1970s, a conical scan air shaker was developed 26 which proved to be quite effective in removing oranges after having been adequately loosened with abscission chemicals. In the early 1980s, two 5-year field studies were conducted with abscission chemicals, air shakers, and trunk shakers. In the first study, harvesting efficiencies of the air shakers averaged 77% while efficiencies of a commercial trunk shaker were 93% in Hamlin oranges and 81% in Valencia oranges. In the second study27 using abscission chemicals with a low frequency (5 Hz) linear trunk shaker, harvesting efficiencies were near 100% in Hamlin oranges and 91% in Valencia oranges harvested prior to a young fruit diameter of 0.5 in.
Robotic harvesting research in the 1980s 28,29,1 developed a single picking arm system with machine vision that could pick 1 orange every 3 to 4 seconds. If perfected, this approach could harvest fresh or processed fruit and selectively harvest mature Valencia oranges. The main problems with robotic harvesting were that only about 75% of the oranges in the tree could be properly identified and not all the identified oranges can be picked because of tree structural hindrances.1,30
In the late 1970s, national federal policy did not support the development of “labor-saving” technology in agriculture. Consequently, USDA funding into citrus mechanical harvesting was dramatically curtailed. Severe freezes of the 1980s significantly reduced Florida citrus production and bearing acreage. The demand or interest in mechanical harvesting research and development waned and R and D projects into mechanical harvesting were phased out by the mid 1980s.
Photo Gallery
*Click blue box in the upper right hand corner of photo gallery to enlarge photo.
-
1990-2008
Written by Jodie D. Whitney, August 2008
After the severe freezes of the 1980s, citrus production in Florida moved to bedded groves in South Florida to minimize the effects of damaging freezes1. In 1991, fruit prices dropped by about one half & many growers could see future production increasing to near previous levels & anticipated that harvesting labor costs & availability could have a significant effect on profitability. An ad hoc citrus industry harvesting advisory committee was appointed to work on harvesting problems. Initially, the industry was more interested in supporting work on labor aids than on mechanical harvesting. In the early to mid 1990s, there were approximately 50 man-positioner boom machines (Harvesting Systems, Ltd., US Patent No. 3,878,957) being operated as a group to harvest oranges for processing from mature trees, & had been under development since the late 1960s.1 Two other machines, the New Way Loader & the Harvest Systems pan machine, were labor aids introduced in the early 1990s. These 2 machines were developed mainly to assist a team of laborers without ladders with harvesting processed oranges from the many young, low-yielding trees which had been planted in South Florida. Even though all 3 machines improved laborer productivity, they did not prove to be economically feasible. A tractor mounted picking aid was briefly investigated in 1994-95, but found that 4 harvesters could not (reach) harvest an adequate percentage of fruit inside a mature tree canopy.
Research in the early 1990s focused on fruit quality & harvester productivity of conventional harvesting methods. Citrus picked into a bag as compared to dropping it to the ground had less decay, defective fruit, trash, microflora, & sand 31. Manual harvesting rates increased by 10% on shorter, productive trees, & increased with increasing fruit yield & fruit weight.32
In 1993, the ad hoc citrus industry harvesting advisory committee organized a harvesting symposium titled “If We Do Nothing” in Lake Wales & invited key industry personnel. Speakers presented information to show that harvesting was a significant problem & asked for industry support to organize a harvesting think tank to develop ideas & guidelines on how to proceed with solutions to the harvesting problem. In 1994, a harvesting think tank was held at the Lake Alfred CREC & the executive summary included several points1, one of which was to hire a harvesting program director. In January 1995, the Florida Department of Citrus (FDOC) hired Dr. Galen Brown as Harvesting Program Administrator & funded the program developed by Dr. Brown & approved by the FDOC’s Citrus Harvesting Research Advisory Council. A sub-committee, the Harvesting Labor Management Committee, worked to improve the safety & productivity with conventional harvesting methods & reported to the Council.
For the 1995-96 harvest season, the Council (FDOC) initially funded four mechanical fruit removal devices to be field tested in processed oranges33. These devices were 2 trunk shakers, an experimental continuous travel canopy shaker, & an experimental canopy penetrator.
Photo Gallery
*Click blue box in the upper right hand corner of photo gallery to enlarge photo.
-
References
Referenced Articles
1A Review of Citrus Harvesting in Florida2Fruit Handling Systems for Florida Citrus
4Engineering Problem in Harvesting Citrus Fruits
5Picking Citrus Fuit By Mechanical Means
6Development of an Augur Picking Head for Selectively Harvesting Fresh Market Oranges
7Citrus Fruit Removal with an Air Harvester Concept
8Harvesting Citrus Fruit with an Inertia Shaker
9Design and Development of a Tree-Shaker Harvest System for Citrus Fruit
10Harvesting Early and Midseason Citrus Fruit with Tree Shaker Harvest Systems
11Performance of Three Air Shaker Patterns in Citrus
12Foliage Shaker for Citrus Harvesting-Part II: Harvesting Trials
13Effects of the Tree Shaker Harvest System of Subsequent Citrus Yields
14Development of a Citrus Removal Device Using Oscillating Forced Air
15Limb Properties of Citrus as Criteria for Tree-Shaker Design
16Optimum Shaking Action for Citrus Fruit Harvesting
17Collecting and Handling Mechanically Removed Citrus Fruit
18Mechanical Removal of Fruit from Citrus Trees
19Shaker Methods for Selective Removal of Oranges
20Biophysical Properties of Citrus Fruit Related to Mechanical Harvesting
21Properties of Young and Mature ‘Valencia’ Oranges Related to Selective Harvest by Mechanical Means
22Chemical Abscission Studies of Oranges and Trials with Mechanical Harvesters
23Abscission Chemicals – Aid to Citrus Fruit Removal
24Shaker Stroke Affects Selective Removal of Valencia Oranges
25Mechanical Systems to Harvest Citrus Fruit for Juice Processing Plant in Florida
26Conical Scan Air Shaker for Removing Citrus Fruit
27A Five-year Study of Orange Removal with Trunk Shakers
28Real-time Vision-servoing of a Robotic Tree Fruit Harvester
29Economic Analysis of Robotic Citrus Harvesting in Florida
30Robotics of Fruit Harvesting: A State-of-the-art Review
31Effects of harvesting practices on damage to Florida grapefruit & orange.
32Orange Grove Factors Affect Manual Harvesting Rates
33Field Test Results with Mechanical Citrus Fruit Removal Devices