An Update on the USDA’s Blueberry Harvest Mechanization Research Project
CREDIT: Fumiomi Takeda: Appalachian Fruit Research Station, U.S. Department of Agriculture, Kearneysville, W.V.
The growth of the blueberry industry in the past three decades has been remarkably robust. For the blueberry industry to remain competitive and sustainable, growers are seeking solutions to ever-increasing problems with labor shortage, and increasingly high labor cost for harvesting blueberries by hand. More and more growers are now using over-the-row (OTR) mechanical harvesters to pick blueberries going for fresh market instead of solely relying on hand pickers. OTR harvesters can reduce the cost of harvesting from over $1.00 per pound by hand to only $0.10 to $0.13 per pound. However, new OTR machines cost as much as $250,000, making high-capacity harvesting machines unaffordable for some small- and medium-size blueberry farms.
Over the years, mechanical harvesters have been used to harvest blueberries going for processing (e.g. frozen and juice). There are 3 main factors to consider for improving blueberry mechanical harvestability: 1) Varieties that have good differentiation in fruit detachment between mature and immature berries, low steminess, high fruit firmness, plant architecture for minimal ground loss, resist bruise damage. 2) Harvest technologies with high selectivity for ripe berry, reduce ground loss, low bruise damage, and high harvest capacity. 3) Production system to facilitate machine harvesting, reduce ground loss, etc. In this article, we have focused on factors two and three.
The USDA got involved in research to develop a harvester designed specifically to pick fresh market blueberry harvesters in early 1990. Eventually, the work resulted in the development of the V-45 harvester made by now defunct BEI, Inc. in South Haven, MI around 1996. It was able to harvest some northern highbush blueberry varieties with long limber canes and the fruit showed little or no physical damage and could be held in cold storage for four weeks or longer. Later V45 trials by a grower in Homerville, GA showed that this harvester performed poorly on conventionally pruned mature Southern highbush and rabbiteye blueberries with stiff, woody branches as it pulled entire plants out of the ground. From 2004 to 2006 USDA and UGA conducted additional research with the V45 harvester on specially pruned southern highbush and rabbiteye blueberry plants. Summer and winter pruning methods were used to remove vertically growing and overarching canes in the center of the bush that removed about 30% to 50% of the canopy to open the middle.
Severe pruning reduced subsequent yields in Brightwell and FL-86-19, but not in Powderblue. On pruned blueberry plants the V45 caused little cane damage. In rabbiteye blueberry, internal fruit damage and skin splitting was less in V45-harvested fruit than in those harvested by a sway harvester and nearly same as hand-harvested fruit. The V45 harvester spread the canes and shook the plants from above and the berries were collected on soft fruit catching surfaces located just underneath the canes. This reduced the drop distance to less than 12 inches as a result less than 20% of fruit harvested by the V45 harvester had damage in more than 25% of sliced surface area showing bruise damage. In contrast, the fruit harvested by a sway harvester >80% of the fruit were severely bruised.
That research and other mechanical harvesting studies have shown that blueberries harvested by conventional OTR machines have significantly reduced shelf life, thus limiting their market distribution. Also, blueberries harvested with conventional OTR harvesters exhibit excessive bruise damage and have lower fruit firmness (g/mm) than hand-harvested blueberries, and their quality deteriorates during postharvest cold storage. Much of quality loss in fruit harvested by OTR machines is from bruise damage. Direct contacts between the fruit and harvester’s hard surfaces cause fruit to become bruised. Hard surfaces on the harvesters like the shaker rods, plastic or fiberglass catcher plates and conveyor belts which collect detached berries, and when the fruit drop from the conveyor belt into the fruit lugs cause fruit to become bruised. These studies have also shown that fruit damage can be reduced by reducing fruit dropping distance and allowing detached blueberries to land on soft fruit catching surfaces.
The blueberry growers in the U.S. and South America are eyeing the lucrative overseas blueberry markets and want a harvester that is capable of picking blueberries that can be held in cold storage for four weeks or more and reach distant Asian markets with high quality. In 2014, a group of blueberry specialists from FL (J. Williamson, S. Sargent), GA (C. Li, H. Scherm, J. Chen), NC (W. Cline), MS (E. Stafne), WV (F. Takeda), PA (A. Freivalds), MI (R. Beaudry), CA (D. Zilberman), OR (W. Yang) and WA (K. Gallardo, L. DeVetter) was successful in obtaining a grant from the USDA NIFA Specialty Crop Research Initiative program to improve mechanical harvesting and handling of fresh-market highbush blueberries. In this multi-state project, the team has been developing and testing harvest technologies and practices that allow for machine harvest of fresh market blueberry with high fruit quality, studying food safety risks associated with conventional and new harvesting technologies, and using sensor technologies to identify factors contributing to bruise damage and also to detect and measure bruise damage in the blueberry non-destructively so that bruised and machine -damaged blueberries can be effectively sorted out on the packing line.
Finally, the group is working to extend the information generated by this project to all blueberry growers. The approach taken to improve mechanical system for harvesting blueberries for fresh market include taking good features of mobile platform, removing bad features that cause damage, develop gentler mechanical harvesting apparatus, change fruit catching surfaces to reduce bruise damage, and use advanced sensor technology to understand harvesting process and nondestructively detect and measure bruise damage within the fruit.
From 2015 to 2017, the group studied a semi-mechanical harvesting system to pick southern and northern highbush blueberries. The focus was to compare hand harvested fruit quality with the quality of fruit harvested by hand-held shakers. To reduce impact damage, the metal and hard plastic catching surfaces were replaced with neoprene-based soft fruit catching surfaces. In Florida, we showed that ‘Farthing’ harvested by this technology were firm with minimal bruise damage and looked as good as hand-harvested blueberries, and the fruit looked quite good even after two weeks in cold storage. However, these harvest-aid platforms lacked the automated, powered fruit conveyance belts to fill lugs. Instead, workers shook the bushes with powered shakers and then manually filled the lugs. The ergonomic analysis of individuals harvesting the bushes with powered shakers indicated that the worker fatigue could become a major issue when working for extended periods.
After studies in FL, CA, OR, and WA with the mobile platform using hand-held shakers, we installed the rotary drum shakers back into the Oxbo 7240, 7440, and 8040 OTR harvesters with modified fruit catching surfaces to increase harvest efficiency and capacity. In 2018, modified 7440 and 8040 harvesters were used to pick ‘Duke’ and ‘Draper’ in Oregon and Washington. In all of these tests, harvested berries were run through commercial packing lines with WECO soft and color sorters or with a WECO color sorter and a MAF Industries optical sorter to obtain fresh pack-out and defect (e.g. green, soft, overripe) data. To sort bruised from non-bruised blueberries among blueberries categorized as fresh pack-out, we developed a non-destructive technique of locating and measuring bruise damage by scanning blueberries with light in the near infrared range. This technology is called hyperspectral imaging and we were successful in identifying blueberries with bruise damage ranging from less than 5% to over 50% inside the fruit.
Our studies in 2018 showed that as much as 92% of blueberries harvested by rotary drum shakers and collected on soft catching surface were sorted as fresh market pack-out. Generally, fresh market pack-out percentage, fruit firmness and extent of bruise damage was slightly to moderately worse than hand-harvested fruit. New sensor technologies will enable packers to sort blueberries into different bruise categories.
2019 WORK PLAN
The high fresh-market pack-out percentage achieved by our prototype blueberry harvesters in 2017 and 2018 were significant improvements compared to commercial OTR harvesters. Currently, we are redesigning the fruit catching surfaces to reduce impact damage and replacing non-food grade neoprene sheets with FDA-listed materials approved for use with food and appropriate for fruit catching surfaces. The plan is to have the redesigned OTR harvester with rotary shakers ready for research in 2019 and compare fruit quality with those harvested by conventional OTR harvesters. We will begin in April in N. Florida by harvesting southern highbush blueberries using a modified Oxbo 8440 and in collaboration with Drs. Jeff Williamson, Steve Sargent, and Patricio Munoz at the University of Florida. We will then move to the West Coast to harvest northern highbush blueberries with a modified Oxbo 7440 in CA, OR, and WA.
Foodborne outbreaks have been linked to the consumption of blueberries contaminated with pathogenic bacteria. The microbial quality of blueberries could be influenced by the conditions of hygiene prevailing during postharvest handling. We evaluated the hygienic conditions of six fresh blueberry packing lines in South Georgia by collecting samples from ten preselected locations on each packing line. Five of the packing lines were sampled twice and the remaining one was sampled once during the summer of 2015 and the summer of 2017. A small area at each sample location was swabbed with sterile sponges before the packing started (AM samples), during lunchtime break (NOON samples), and at the end of a packing day (PM samples). The sponges were sampled for total aerobes, yeasts and molds, total coliforms, fecal coliforms, and enterococci. The results showed that sample site and sampling time had a significant influence on total aerobic, yeast and mold, and total coliform counts. The PM samples had significantly higher total aerobic and yeast and mold counts than the NOON samples which had significantly higher counts than the AM samples. Forty-six out of the 310 (14.8%) collected samples tested positive for enterococci while 27 (8.7%) samples tested positive for fecal coliforms. Berry lugs, rubber belts on color sorters, and premature berry disposing areas had significantly higher microbial counts than the other sites. The study suggests that some sites along fresh blueberry packing lines could become contaminated by microorganisms during packing.
Whether these contaminated sites will become a food safety concern depends on the incidence of pathogen presence in the microbial community and efficacy of routine sanitizing treatments. A few sites such as rubber belts in the color sorter area, immature berry disposing area, and berry lugs require special attention in terms of maintenance and sanitation treatment. Certain recommendations could be made to improve the hygiene conditions of blueberry packing lines. For example, i) properly prepare and timely replace sanitizer solutions, ii) routinely clean and sanitize berry lugs and packing lines, iii) follow standard sanitation practices to keep the level of microbial counts on sanitized surfaces as low as possible, iv) keep berry contact surfaces in good condition, free of cracks and other physical defects, and v) frequently inspect packing lines during packing operations for residue build-up or cleaning needs. The efficacy of four sanitizers (ozonated water, chlorine dioxide, quaternary ammonium compound, and sodium hypochlorite) for removing biofilm tested on coupons made of materials commonly used on fresh blueberry packing lines, lugs and clamshells, including stainless steel, rubber, and PVC, polyurethane, HD polyethylene and polyethylene terephthalate were evaluated.
In summary, biofilms accumulated in greater amounts on polyethylene coupons than on all other coupons used in the study. Ozonated water was more efficacious in removing biofilms than quaternary ammonium compound, which had higher efficacy than chlorine dioxide and sodium hypochlorite. Overall, rubber surface was more difficult and stainless-steel surface relatively easy to sanitize. These results emphasize the importance of selecting proper chemical sanitizers and blueberry packing line surfaces because these decisions will affect the hygiene conditions of packing lines as well as the microbial quality of the blueberry fruit.
More details about our research project can be found in the publications listed below.
Gazula, H., Quansah, J., Holland, R., Scherm, H., Li, C., Takeda, F., Chen, J. 2019. Microbial loads on selected fresh blueberry packing lines. Food Control. 100:315-320.
Takeda F., Yang, W.Q., Li, C., Freivalds, A., Sung, K., Xu, R., Ho, B., Williamson, J., Sargent, S. 2017. Applying new technologies to transform blueberry harvesting. Agronomy. 7(2):33.
Yu, J., Takeda, F., and Li, C. 2016. Nondestructive detection and quantification of blueberry bruising using near-infrared (NIR) hyperspectral reflectance imaging. Scientific Reports. 21(6):35679.
Wasko DeVetter, L., Yang, W.Q. Takeda, F., Korthuis, S., Li, C. 2019. Modified over-the-row machine harvesters to improve northern highbush blueberry fresh market quality. Agriculture. 9:13.