Developing Hazelnut Germplasm for the Upper Midwest

Project Overview

Project Type: Research and Education
Funds awarded in 2010: $175,000.00
Projected End Date: 12/31/2013
Region: North Central
State: Minnesota
Project Coordinator:
Dr. Lois Braun
Research Associate

Annual Reports


  • Nuts: hazelnuts


  • Crop Production: windbreaks
  • Energy: bioenergy and biofuels
  • Natural Resources/Environment: afforestation, biodiversity, hedgerows, riparian buffers, hedges - woody
  • Production Systems: permaculture


    Hybrids between wild native American hazelnuts (Corylus americana) and domesticated European hazelnuts (Corylus avellana) combine the disease resistance and winter hardiness of the former with the large nut size and nut quality of the latter. As a woody perennial shrub, they are a potential “Forever Green” crop for the Upper Midwest, offering multiple ecological benefits while producing a tasty and healthful human food with high market demand. The primary obstacle to their wide-scale adoption is lack of reliable improved germplasm. This project starts to address that by taking the first steps to build a breeding program.

    We have now collected 157 accessions of hazelnuts (129 hybrids, 22 C. Americana, and 6 C. avellana) which are now represented in five hybrid hazelnut germplasm performance trials in the Upper Midwest (St. Paul, Lake City and Lamberton, Minnesota, and Bayfield and Tomahawk, Wisconsin). The oldest of these have been bearing nuts for three years now, so we are starting to identify the best of these to take to the next steps. 1) Some of the best have been cross pollinated with advanced selections of European hazelnuts from Oregon. About 550 seedlings of these crosses have been planted to the field and more seed from crosses are being germinated. 2) The best four selections will be micropropagated for dissemination to growers for further evaluation and to use for agronomic trials.

    Dissemination of improved woody plants depends on efficient methods of vegetative propagation to develop the large numbers of clones needed. Mound layering, which was used to produce the plants in the performance trials, cannot produce the numbers needed. So we have been working to develop methods of rooting hardwood stem cuttings as an alternative. We just completed a how-to guide on the method, and have written a series of scientific papers on this research that we hope to publish in Hort Science soon. The method is not reliable enough for mass propagation, but it can produce enough plants for research purposes in situations where mound layering is not feasible. For mass propagation we will need micropropagation, which is also being developed, with different funding.

    Even as we seek to improve germplasm, growers who have planted seedling stock or hope to do so soon, need advice on how to plant and manage them, so we have also been working to address their questions. Many early growers complained of poor transplant survival. So we conducted trials to determine the best kind of planting stock and the best planting time. We concluded that vigor of planting stock is imperative, and that vigor is related to a combination of pot size and age of plant. Healthy bare-root dormant stock can be planted in either fall or spring, whereas non-dormant planting stock should be planted in late summer or fall.

    With this grant, we also established controlled and replicated trials to compare methods of pre-plant ground preparation, weed control, and N fertilization, using clonally-produced plants that had not been available before this grant. Earlier attempts to address these had failed because of the high level of genetic variability within the trials. Although these trials are still too young to have produced conclusive results, tentatively we recommend eliminating all perennial weeds during pre-plant ground preparation, and maintaining good weed control through the first two years of plant growth by any of a variety of methods.

    For outreach we have organized an annual Upper Midwest Hazelnut Growers Conference, hosted field days and given winter seminars. For more information see our website at


    The long-term goal of this project is to develop native hazelnuts (Corylus americana) and their hybrids with European hazelnuts (C. avellana) as alternative crops for the Upper Midwest, for the purpose of diversifying agriculture to enhance ecological and economic sustainability. Hazelnuts are the only woody perennial amongst the “Forever Green” crops being developed at the University of Minnesota. Ecological benefits of woody perennials include covering the soil and protecting it from soil erosion year round; reducing runoff and leaching, and thus protecting water quality both from sediments and nutrients; increasing soil carbon sequestration and enhancing soil quality; and providing habitat for wildlife. Hazelnuts can also stabilize agricultural incomes because they can provide economic return from windbreaks, shelterbelts, living snow-fences, lake and stream-bank buffers, and marginal land. They are more resilient than annual crops because they are less sensitive to drought and flooding, they have lower requirements for tillage, fertilizers, and pesticides, and timeliness of most of their management practices is not critical.

    The multifunctionality of hazelnuts makes them a potential foundation species for a perennial plant-based bioeconomy. They have a four-tiered potential market: growers may sell them high-value fresh-eating nuts, as ingredients in processed foods, or for pressing into oil, and may also collect payments for the conservation services they provide. Hazelnuts are a healthful food, being high in healthful monounsaturated fatty acids, vitamin E, thiamin and fiber (Richardson, 1997). Hazelnut oil has the same healthful fatty acid profile as olive oil, and can be used in all the same ways, including in cosmetics and skin-care products. Hazelnuts also have potential as a biofuel because their oil quality is superior to soybean oil for biodiesel and because their oil content is 50 to 70% oil, about three times as high as soybeans (Xu et al., 2007). Although with current genetics hazelnut yields could not produce as much oil per acre as soybeans, individual plants evaluated in Nebraska would, if mass propagated, be able to produce three times as much as soybeans (Hammond, 2006).

    The major obstacle separating this potential from reality is lack of consistent highly performing germplasm. The hybrid hazelnut seedlings available in this region descended from crosses made between Corylus americana and Corylus avellana made in the 1930s by amateur breeders to combine the disease tolerance and winter hardiness of the former with the superior nut size and quality of the latter. But in spite of breeding efforts involving multiple generations of recurrent selection, the planting stock currently available is still almost entirely seed-propagated from open-pollinated stock. This means that that the superior genetics responsible for the outstanding yields found in some individuals are not consistently passed on to their progeny. A few progeny may be like their parents or better; however, most will be inferior. It is impossible to establish hybrid hazelnut plantations that are profitable with this kind of inconsistency and unpredictability. Moreover, evaluation of parent material has not been replicated, and thus parents assumed to have outstanding characteristics may have been selected due to locally favorable environmental conditions rather than due to their genetics.

    This grant is the start of efforts to establish a long-term hazelnut germplasm improvement program for the Upper Midwest. Our first steps were to identify superior hybrid hazelnut germplasm from the region’s farms, and superior wild American hazelnut germplasm from its natural areas, propagate it vegetatively with mound layering, and gather it into five replicated performance trials. This enables us to distinguish between the effects of genotype and environment, and to identify germplasm that is truly desirable. We are now moving on to the next steps—cross pollinating the best genotypes with advanced selections of European hazelnuts from Oregon, and propagating them for field scale trials, and possible eventual release as improved varieties.

    Concomitant with improving germplasm, we have been working to develop more efficient methods of vegetative propagation, which will be needed to disseminate the improved varieties that we develop. The mound layering technique we used to establish our germplasm performance trials are capable of producing only a few dozen new plants per year from each mother plant. With this grant we investigated how to root hardwood stem cuttings as an alternative. This grant also enhanced our ability to secure funding to research micropropagation, which is potentially a much more productive method.

    Additionally, this grant funded agronomic trials to answer some of the questions dogging the 150 some hazelnut growers we have identified in Wisconsin, Minnesota, Iowa and Illinois. We tackled the poor transplant survival of hazelnut seedlings reported by many growers with trials comparing type of transplant stock and transplant timing. We also established trials to compare methods of pre-plant ground preparation and post-plant weed control. Many hazelnut growers (who are frequently small landowners, not full time farmers) would like to plant hazelnuts into former CRP or other highly sloping land with sod ground cover, but are not equipped to till it up, nor want to because of concerns about soil erosion and soil ecology. This led us to test no-till planting of hazelnuts, even though this presents weed control challenges. Also, many growers have had the impression that weed control is not essential for hazelnuts, or else lack equipment for conventional methods of weed control, so we established trials to demonstrate both its importance and some methods that might be used. These trials will also enable us to calculate the economic return on various weed control methods.

    This grant has also funded outreach events, in the form of conferences, seminars, field days and a website, to get information about improved practices out to current growers, prospective growers and the public at large. These are also important for maintaining grower enthusiasm during the long wait for improved germplasm.

    Hammond, E.A. 2006. Identifying Superior Hybrid Hazelnut Plants in Southeast Nebraska. Master’s Thesis. University of Nebraska.

    Richardson, D.C. 1997. The health benefits of eating hazelnut: implications for blood lipid profiles, coronary heart disease, and cancer risks. p. 295-298 In: A.I Koksal, Y.

    Xu, Y.X., M.A. Hanna, and S.J. Josiah. 2007. Hybrid hazelnut oil characteristics and its potential oleochemical application. Industrial crops and products 26:69-76.

    Project objectives:

    1. Plant Material
    • We exceeded our target of four germplasm performance trials, with 100 accessions in each, by establishing replicated five performance trials, with 85, 92, 114, 119 and 157 accessions respectively, in triplicate to the extent possible. Of the 157 accessions represented in the largest performance trial, 22 are Corylus americana, 6 are Corylus avellana, and the remaining 129 are hybrids between the two species.
    • We produced enough clonal plant material, through either mound layering or hardwood stem cutting propagation, to establish four N fertilization trials (in 2011 and 2012), four ground preparation/weed control trials (in 2013), and one plant spacing trial (in 2014). These trials are still too young to have yielded results, but results may start coming in as early as fall 2015.

    2. New Knowledge
    A. Commercially viable methods of hazelnut propagation
    • Three papers detailing our hardwood stem cutting trial have been written and will soon be submitted to Hort. Science for peer review and publication.
    • We developed a how-to bulletin on propagation of hazelnuts from hardwood stem cuttings using low-cost humidity tents. The bulletin will be posted on our website at and also on the SARE website. Although this method falls short of being commercially viable, it is useful for producing clonal plant material for research purposes in situations where mound layering is not viable.
    • Our trials to improve rooting success with mound layering by figuring out how to best manage stock plant beds were inconclusive. 1) We found no differences in rooting between stock plants that were coppiced for layering after four versus three years, and 2) we found no differences in rooting between fertilized and unfertilized stock plants. We do not plan to publish this work because the experimental design was faulty.

    B. Management of clonally propagated planting stock
    • Type of planting stock. New information from our trials comparing different types of planting stock and different transplanting dates was incorporated into an updated version (Version 3) of our Hazelnut Production Guide, available at Briefly, we concluded that healthy bare-root dormant planting stock can be successfully transplanted in either spring or fall, though non-dormant planting stock is better transplanted in the fall. This applies to both seedlings and clonal planting material. Plants produced by mound layering can safely be left in the mound until spring if the fall transplanting window is missed. The most important determinant of transplant survival is the quality of the young plant; the best quality plants are grown with enough space for vigorous root development. Transplant survival of first year layers that were grown in a nursery bed for a year before transplanting to their final location is high, mostly because weak layers are winnowed out before transplanting, but this is offset by higher transplant cost and labor.

    C. Establishment of hybrid hazelnuts
    • Pre-plant ground preparation and early establishment weed control. We were overly optimistic including this trial in our 2009 proposal because we were not able to get these trials planted until we had generated sufficient quantities of clonal planting stock for them. So the weed control trials didn’t get planted until 2013, which means that it is too early for conclusive results. Subjective impressions support the advice that good weed control is important for vigorous early growth of hazelnuts, but there appear to be no differences between methods of weed control. Choices about methods will depend on the resources available to growers and their personal preferences. However, when the bushes in these trials start bearing nuts, so that we can calculate an economic return on them, some weed control methods might emerge as being more cost effective than others.

    3. New Knowledge Shared with Growers and Other Researchers
    Outreach activities are listed in the section on outreach.

    4. The Second Upper Midwest Hazelnut Growers’ Survey was completed in Nov. 2011: A third growers’ survey in early 2014 was not completed because of poor response. Perhaps our committed growers feel like they’ve already answered our questions, leaving the growers who aren’t committed enough to respond to surveys.

    Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.