Final report for FW18-048
Nearly all hazelnuts produced in the United States are grown in Oregon’s Willamette Valley.
Hazelnuts are typically harvested by sweeping them off the orchard floor every fall. This “sweep
harvest” method necessitates keeping orchard floors very flat, and free of vegetation/debris that
may interfere with harvest.
Keeping orchard floors “clean” is achieved through frequent flail mowing, leveling and herbicide
applications. These practices can lead to soil degradation. Additionally, because sweep
harvesting cannot begin until all nuts have fallen from the tree, harvest is often conducted after
the onset of the fall rains which can lead to additional compaction and difficulties separating the
nuts from debris and mud.
We are proposing to study the potential for shake and catch harvesting in hazelnut production. If
feasible, shake and catch harvesting could allow for a whole host of environmental benefits as
well as increase the economic viability of small to midsize hazelnuts orchards.
With shake and catch harvesting techniques, nuts are harvested straight from the tree by shaking
the trunk, causing nuts to fall onto an inverted umbrella and into a tote. This would allow for
much less intensive orchard floor management, as nuts do not have to be swept off a clean and
manicured orchard floor. Deep-rooted cover crops could be grown to maximize their soil
building potential, and animals could be integrated after the harvest to help control disease and
maintain fertility. Finally, harvest could happen earlier in the fall, decreasing the potential for
harvesting in the rain.
In order to test the viability of shake and catch harvesting, we are proposing to compare yield,
and timing of “shake and catch harvesting” vs “sweep harvesting” in order to assess whether
shake and catch harvesting has the potential to replace traditional sweep methods in hazelnut
production in Oregon and beyond.
We plan to share our experience and findings with the wider agricultural community through
online demonstration videos, a dedicated page on our website and regular updates via our
Facebook page, Instagram account and email list. We plan to host two field days on our farm and
to share our findings at the Oregon Small Farms Conference.
1) Compare yield of shake and catch to conventional harvest
a. We want to know if we can harvest as many nuts with a “shake and catch
harvester” as with a “conventional sweep harvester”. We will be measuring
yield by finding the percentage of total nuts we are able to harvest from a
given block of 118 trees (one acre). We will then compare our harvest
percentage to the industry average.
2) Compare quality of nut in shake and catch to conventional harvest
a. Conventional sweep harvesting waits for hazelnuts to fall to the ground at
maturity. Shake and catch harvesting will be conducted earlier in the fall
before nuts drop to the ground. Therefore, we want to know whether shake
and catch harvesting has any effect on the ability of the hazelnut to fully fill
its shell. We will be measuring quality much in the same way that major
hazelnut processors do: by looking at percentage of nut fill. We will crack
and separate nuts from shells for our given block of 118 trees. Then we will
divide weight of shell by weight of nut to get our nut fill percentage and
compare that to the industry average.
3) Compare percent blanks in shake and catch to conventional harvest
a. We want to know if shake and catch harvesting has any effect on the
percentage of nuts harvested that are blanks (have no nut inside). We’ll test
this much the same way that large processors do, by cracking random samples
of nuts from our test block one at a time and recording the percentage that are
blank. We will then compare this to the industry average as well as to other
blocks in our orchard.
4) Compare harvest timing of shake and catch to conventional harvest
a. Shake and catch harvesting has the potential to occur and be completed earlier
than sweep harvesting methods. We plan to document actual and preferred
harvest dates in 2018 and 2019 and then compare those with the industry
average harvest dates for those years.
- (Educator and Researcher)
In our orchard of just over 2000 hazelnut trees, we randomly selected 138 to test for yields. 69 (half of the trees) were harvested by the traditional method of hand-picking. The other half were harvested via shake and catch with an umbrella harvester on a single pass. For trees we harvested via shake and catch, we made a second pass using the traditional method so we'd be able to determine what portion of the nuts were caught. We bagged each tree's harvest individually to take to the lab.
In the lab, we weighed each bag of nuts. This data revealed the yield from shake and catch. We then took 3 random nuts from each bag. Each nut was weighed, cracked, and compared to the weight of its kernel. From this we were able to determine nut fill, the ratio of kernel wight to in-shell hazelnut weight. We then compared nut fill in the shake and catch method to the traditional method of harvesting.
We used this method in the fall of 2018 and 2019
By comparing nuts caught to nuts picked up afterwards, we determined that in 2018 62.4% of hazelnuts were picked up in a single pass with the shake and catch method in 2019 we picked up 65% of total nuts giving us an average harvest of 63.75% of nuts harvested with the Shake and Catch method. Compared to expected yields from current industry standard (estimated 80-85%), this is pretty low.
By cracking out individual nuts from each tree, we gleaned quite a bit more information. One of our hypotheses was that shake and catch would yield fewer blanks (nuts without kernels). We thought this would take place because blanks are notorious for dropping early. The data did not support this as each method yielded 13% blanks in 2018. We used this same method in 2019. Traditional harvest methods yielded 65% blank whereas shake and catch yielded 23% blank. This is higher than we would expect but due to stressors in the orchard included a late fall drought in 2018 which led to low yields overall.
Additionally, we calculated the nut fill per tree for traditional and compared it to shake and catch. In 2018 Nut fill on traditional was calculated at 34.8% whereas shake and catch gave 38.7%. In 2019 nut fill for traditional harvest was 15.8% whereas shake and catch nut fill was 37.6%. Thus our average over the two year experiment was 25.3%% nut fill for traditional harvest and 35.15% nut fill for shake and catch harvest. This 10% difference is interesting and may be due to the fact that squirrels are carrying off nuts from these young trees that spend much time on the ground whereas they aren't as able to get at them in the trees. We also believe the lower nut fill is due to the fact that the first nuts to drop are the ones that drop prematurely, before filling fully.
Education and Outreach
Attended the Nut Grower's Society meeting representing the Oregon Organic Hazelnut Cooperative (OOHC) and promoted this project at our table.
We had several members of the the OOHC come and watch us use the Olinet.
We posted a video of using the Olinet on our website and social media.
We included a description of the project in our fall newsletter.
We included a tour and discussion of the project at our Fall Festival which ran every Saturday in October 2019 and was attended by over 500 individuals.
Education and Outreach Outcomes
We feel that Shake and Catch harvesting on it's own is probably not a viable option for commercial hazelnut producers at this point. However, as we mentioned above, continuing to explore ways to further integrate livestock and biological diversity into hazelnut orchards is an area of research that is ripe for innovation.
Became aware of the potential for using shake harvesting for hazelnuts and other tree crops.
Gained skills in Excel data management.
Increased openness to using novel harvest techniques for hazelnuts.