Developing Hazelnut Germplasm for the Upper Midwest

Developing Hazelnut Germplasm for the Upper Midwest

Summary

The oldest plants in our five hazelnut germplasm performance trials (which now have up to 121 accessions in triplicate at each) are starting to bear nuts, so it will not be many years before we are able to make our first selections. We have completed two years of scouting for wild Corylus americana on public lands and will soon start adding exceptional American hazelnuts to the performance trials. We have also made our first controlled crosses and are now planting our first F1 seeds. A graduate student is starting research on micropropagation of hazelnuts. After determining that supplemental light is beneficial for rooting hazelnut stem cuttings, this year we are experimenting with alternative means of maintaining humidity while avoiding overheating in the rooting tents. We have established four new plantings of clonal material for N fertilization trials, initiated three pruning and coppicing trials, and have produced the plant material for four weed control trials to be planted in the spring of 2013.

Objectives/Performance Targets

Outputs
1. Plant material
• Five germplasm performance trials expanded to at least 100 plant accessions in each.
• Seeds of the first F1 population from crosses between local hybrids and European selections from Oregon are currently germinating in pots in the greenhouse.
• Four new plantings of clonal material have just been established for N fertilization trials and enough clonal material has been produced to plant four weed control trials in 2013.

2. New knowledge is being developed on the following, with bulletins on each to be posted on the hazelnut website:
• Commercially viable methods of hazelnut propagation.
• N fertilization recommendations for mature hybrid hazelnuts.
• Management of clonally-propagated planting stock.
• Weed control methods during establishment.

3. As new knowledge is developed it is shared with growers and other researchers via the Upper Midwest Hazelnut Development Initiative website, summer field days and an annual regional hazelnut conference.

Accomplishments/Milestones

Objective 1. Germplasm Improvement.
In 2012 we added seven new accessions to our five germplasm performance trials (St. Paul, Lamberton, and Lake City, Minnesota, and Bayfield and Tomahawk, Wisconsin), bringing the number of accessions represented up to 121. Of these 106 are hybrids of Badgersett stock, identified by growers from their plantings; 11 are hybrids of non-Badgersett stock, either purchased from other nurseries or layered on the farms of growers who purchased from those nurseries; and 4 are C. americana. The rate of addition of new hybrid hazelnut accessions is slowing, as we have already included most of the outstanding bushes identified from our most mature plantings, though a few more will be added as newer plantings start to mature and as new outstanding bushes are identified by growers participating in the Hazelnut Improvement Program (http://www.midwesthazelnuts.org/about-hip.html).

As the addition of hybrid hazelnuts accessions slows, the addition of C. americana is picking up. After two years of scouting for superior wild American hazelnut germplasm on public lands, we have now identified one or two candidates at each of 14 locations. In November 2012 we coppiced 22 of these bushes for propagation in 2013, either by mound layering or from hardwood stem cuttings. Mound layering at wild sites will be challenging because of the distance and difficult terrain over which large quantities of sawdust will need to be hauled, and because of the potential for disturbance by wildlife and unwitting humans. Thus we will propagate them primarily by stem cuttings. This work has already begun because at three of our scouting sites we found that bushes identified as candidates in 2011 had either been mowed or burned down in the course of prairie management in 2012, meaning that we were unable to assess them for a second year. However, all of these bushes had vigorous regrowth and produced abundant crown suckers of the type that are ideal for hardwood stem cuttings, so we went ahead and collected them even though we had only one year of data on them. In addition, we were able to collect a few collar suckers from some plants that had not been mowed or burned. Stems were thus collected from a total of 14 accessions of C. americana and were placed into hardwood stem cutting trials in February 2013. (Two of these C. americana bushes supplied so many uniform stems that we were able to replicate the hardwood stem cutting trials to a greater level than ever before, greatly improving our chances of getting publishable results.) The six outstanding C americana from which we were unable to collect stems in fall 2012 will be mound layered in 2013 if they are close to a road; otherwise we will collect stems from them for stem cutting propagation in 2014. In addition, we will continue scouting for additional plants at additional sites.

Our colleagues in Wisconsin, Jason Fischbach and Mike Demchik, are taking a slightly different approach for propagating the approximately 20 C. americana that they have identified in the wild: recognizing the difficulty of mound layering in the wild, they dug these plants up, divided them in half, and planted half in a nursery in Bayfield and half in pots. These plants will be the source of germplasm for micropropagation in the lab of Brent McCown at UW-Madison.

Although a handful of the oldest plants in the germplasm performance trial yielded nuts in 2011, in 2012 they bore their first significant nut crop. Yield and nut quality evaluations of these are currently underway in the lab.

In 2012 we got our first confirmation that field inoculations for screening for resistance to Eastern Filbert Blight (EFB) have been successful, with the observation of one infected plant each in St. Paul and Lake City. Before this we had not been confident of our inoculation method. (In other older trials we have not observed EFB until the planting is a decade or more old.) Now we can expect the disease to spread readily to other susceptible plants in the plantings, enabling us to screen for resistant plants.

In 2012 we also made our first controlled crosses. We had not planned to do them this soon, because we did not anticipate having sufficient data with which to select mother plants this soon. Our purpose was merely to practice doing controlled pollinations and handling the resulting seed, so that we know how to do so when the time comes to scale up. Thus we used as maternal parents four plants from early stem cutting propagation work which were un-replicated and thus which were of unproven value, but which had displayed desirable nut characteristics the previous two years. We used pollen from seven European varieties developed in the hazelnut breeding program at Oregon State University, four of which have EFB resistance from the Gassaway gene and one with resistance from another source, for a total of twelve crosses. The resulting seed from the 2012 crosses are currently germinating in the greenhouse and will be screened for EFB resistance using the methods of Coyne et al, 2000 and Molnar et al., 2005. In spring 2013 we plan make the remaining 16 crosses needed to complete a 7 by 4 diallele, for a total of 28 crosses.

Objective 2. Propagation Methods.
Hardwood stem cutting trials. After two years of hardwood stem cutting trials, in which we never attained rooting rates of higher than 24%, we hoped for a breakthrough in our third set of trials in winter 2012, made possible by combining everything we had learned in the previous two years. We had learned that the ideal concentration of IBA is about 2,000 ppm. We had also learned that the best rooting is from relatively small first year crown suckers that are not cut into shorter segments, but that, if they have to be cut into shorter segments in order to fit in the rooting chambers, it is best to keep them as long as possible, presumably because larger stem segments have more stored carbohydrate with which to grow roots. We had also learned that starting them in early February is better than later, presumably because then the stems are not subjected to the extreme heat that can build up in the greenhouse in late spring and summer.

The objective of our 2012 trials was to determine why our attempt to replicate the system in two of our growers’ greenhouses had failed. Our hypothesis was that it was due to a combination of low temperature and light. We confirmed that supplemental light via Halogen greenhouse bulbs is indeed beneficial, but our results about temperature were ambiguous. They suggested that higher temperatures are beneficial early in the season but harmful later in the season, when lack of ventilation in the sealed humidity tents resulted in temperatures of up to 1250 F.

Although we did not find the breakthrough we were looking for to enhance rooting rates, in 2012 we radically improved survival in the transition from hothouse to pots in an outdoor nursery. Previously we placed the newly rooted stems back into a high humidity tent under the assumption that they were not adapted to the lower humidity outside conditions and gradually hardened them off over a period of about three weeks. In 2012 we accelerated the hardening off period, starting it within a few days of transplanting, and completing it within two weeks. This resulted in 92% survival of rooted cuttings to field transplanting, presumably by sparing the young plants the heat stress that occurs in the humidity tents. Survival after field transplanting has generally been good, up to 100% for the most vigorous genotypes.

Our 2013 trials, which were implemented in January and February, are testing whether this heat buildup can be avoided by using ventilated humidity tents, in which humidity is maintained with humidifiers. By implementing part of the experiment in January, a month earlier than in the past, we are testing another approach to avoiding heat build-up. Because the stems placed in January have broken bud we are confident that this is not too early. An undergraduate enrollee in the Undergraduate Research Opportunities Program is helping with this project.

Another undergraduate is helping with a trial to determine whether rooting might be enhanced by keeping the rooting chambers in a cooler on top of a heating mat: the mat keeps the root zone warm, thus stimulating root growth, while the cool air temperatures inhibit shoot growth, limiting the stress that shoot growth places on these rootless stems. We are also testing whether the addition of 1,000 ppm of NAA, a synthetic auxin that is more stable than IBA, to our IBA solution will enhance rooting.

For the first time in our hardwood stem cutting trials we have had large numbers of uniform stems to work with, some of them from C. americana from sites that had been burned for prairie restoration. This has enabled us to replicate our trials at the levels we need to be able to publish them.

Stock plant beds/mound layering trials. Plants produced by mound layering in 2008 were finally large enough in 2012 to themselves be mound layered, so we were able to implement our first truly replicated layering trials with them. However, our enthusiasm for this trial dampened when we realized that those plants that were old enough for it, which were themselves the result of layering trials on unselected plants, were not especially desirable. Their only positive attribute is that they propagate easily, like lab rats, and consequently, the trial would not generate the abundant clonal material we needed for our agronomic trials.

Our four years of observations on individual layered plants led us to the following hypothesis about why mound layering is sometimes productive and sometimes not: assuming the mother plant has genetics amenable to propagation, mound layering will be more successful when the mother plant has adequate nutrition, moisture and sunlight. Consequently, mound layering will be more successful when the mother plant has grown large enough to have a well developed root system, provided that it is not shaded by its neighbors. To test this we mound layered half of the plants of each of four genotypes in 2012, leaving the other half to grow another year before being layered in 2013. We then applied a slow release N fertilizer to half the plants in each group to test whether supplemental fertility would be beneficial. (Soil tests indicated that N was the only nutrient likely to be limiting.)

Even though we initially established the stock plant beds for the purpose of providing clonal material for highly replicated mound layering trials, in 2012 we decided that these plants, most of which were planted in fall 2009, would have more use in our agronomic trials. Thus in October 2012 we dug them up and transplanted them out as described in the section on agronomic trials below.

Micropropagation. Our collaborators at UW have reported good results micropropagating wild hazelnut germplasm, while at the University of Minnesota we have a new graduate student, Molly Kreiser, tasked with figuring out why genotype is usually the strongest determinant of success in micropropagation, as it is in all methods of vegetative propagation. By figuring this out we hope to figure out what can be done to propagate all genotypes. We are hopeful that micropropagation will prove to be successful enough as to eliminate the need for other methods of propagation. However, the stem cutting research is still useful because micropropagated shoots must eventually be transferred into a greenhouse for rooting by much the same methods.

Objective 3. Development of Best Management Practices for Establishment.
N fertilization trials. In our initial proposal we discussed the need for N fertilization trials using clonal material, to simplify the statistical analysis of the results, which in earlier N trials (Braun, 2008) had found to be complicated by the extremely variable responses due to the genetic variability of seed-propagated plants. There is also a need for fertilization trials on mature bearing hazelnuts because, whereas the earlier trials had determined that the N requirements of young non-bearing hazelnuts is very low, N requirements are likely to rise as N is exported with the nuts that are sold off the farm.

We had thought that we would need to wait to begin N trials until we had produced enough clonal material with which to do them. But in 2011 two growers made available to us their plantings of mature clonal stock. We have now completed two years of N trials on these. In addition, we have completed two years of N trials on the mature open-pollinated planting at the Staples Ag Center, where we figured that five years of background data to use as a covariate in data analysis would make up for the genetic variability. First year results at all sites showed leaf N responses to fertilization, but no yield responses, which is consistent with earlier observations that in woody plants there is a lag time before yield responses are observed. Second year data has been collected but has not yet been analyzed.

There are challenges at all of these sites: one of the growers has objectives that are in conflict with our experiment; the genotype of the other clonal planting is not very productive, which may make the results irrelevant; and the planting at Staples is succumbing to EFB. Thus we have moved ahead with our original plan to establish new plantings for these N trials, using our own clonal material. In the fall of 2011 we established two new plantings for future N fertilization trials, using 2011 mound layers, and in the fall of 2012 we established an additional two new plantings, using plants dug up from the stock plant beds, most of which were layered in 2009. These trials were designed with three plants per plot (all three of which will receive the same N treatment), and ten treatments (five N rates in a factorial with and without another nutrient, either P or K, depending on what is likely to be limiting at the site), and from three to six replications (depending on available plants and space). The two 2011 plantings (at Rosemount and on a farm in Fillmore County) have 6 and 3 reps (180 and 90 plants) respectively, and the two 2012 plantings (at Staples and Becker) also have 6 and 3 reps respectively. The plantings at Rosemount and Fillmore will not be ready to start implementing N treatments until they are three or four years old, but the plantings at Staples and Becker, which used plants that were already up to three years old will start to be fertilized after they have started to yield consistently, which may be as soon as 2014. These trials are maintained with a high level of weed control, using a combination of woodchips and spot applications of Roundup.

Pruning and Coppicing Trials. Since 2005 we have been collecting yield data on three hazelnut plantings, established in 2000. These plantings now need to be rejuvenated. They are becoming overcrowded, and the plants are now too tall to pick easily by hand, and likely too tall to pick by machine as well. Yield on two of them was dramatically lower than usual in 2012. This could have been due to the drought of late 2011, or due to the typical alternate year bearing habit often observed in woody crops. Or it could have been due to an excess of older unproductive wood. Badgersett Research Farm advocates rejuvenating hazel bushes by coppicing them to the ground when they reach this stage. Although coppicing completely eliminates yield for a year and reduces it for a second year, subsequently yields return to previous levels or greater. But might not annual renewal pruning work as well, without loss of a year of yield?

In spring 2012 we initiated pruning trials on two of the plantings described above (at Rosemount and Staples), as well as on a slightly younger planting established in 2003 (at the Horticultural Research Center at Chanhassen), to find out. Two intensities and two systems of pruning were compared with coppicing and the control, which was no cutting at all, for a total of six treatments. The response evaluated will be three or four year cumulative yield, with baseline yield as a covariate. Thus results will not be available for three years. We can, however, say at this time that pruning requires significantly more labor than coppicing. We can also say that the plants did not respond to pruning in a negative way, as we feared they might.

One of the challenges of this trial is that there are so many approaches to pruning that we were not sure which ones to test. In 2013 we will initiate the trial on the third older planting, slightly better informed. In 2013 we will also initiate pruning trials on some three- or four-year old clonal plantings, because it is likely that pruning works best on younger plants.

Ground Preparation and Weed Control. In 2012 we finally generated enough clonal planting material for the long awaited weed control trials, but we ran out of time to get them into the ground before winter set in. The plants, all of them 2012 mound layers, are currently heeled into a nursery in St. Paul and will be transplanted to four paired trials at Rosemount and Waseca just as soon as the ground is workable in spring 2013. There will be two sites at each location: one will be a row crop field, which has been recently tilled, and the other a grassland, which has not. These represent the two most common starting conditions for growers planting hazelnuts, and each will present a unique set of weed management challenges. We have not yet finalized the weed management treatments to be used, because there are too many options to try them all, especially when considering that intensity of management is likely to be more important than method of management. For example, a one inch thick layer of woodchips extending six inches around the base of a transplant is likely to be worthless, but three inches for three feet is likely to be very effective, if the grower can procure the needed woodchips. Likely treatments will probably include the following: woodchipping, landscape fabric, hoeing, herbicides (pre and or post-emergence), mowing and nearly complete neglect. Our objective is not to discover what methods work, which we already know, but what the plant response will be to different intensities of management, which is essential information for calculating the cost effectiveness of different methods.

Other trials. Surplus plants generated by the 2012 stem cutting trials and from miscellaneous other sources make possible a few other small trials that we were previously unable to do:
• Tree tubes. In fall 2013 a small trial at Rosemount was implemented to test the value of Plantra ® tree tubes in the establishment of hazelnuts. We had previously assumed that tree tubes would not be useful for hazelnuts because their bushy growth form would be confined by them, or they would send suckers up outside of them. But Plantra now has available some wide tree tubes, which we are comparing with their traditional narrow ones. In the spring we will repeat the experiment in St. Paul, and also look at the effect of half-high tubes.
• Herbicide tolerance. Some growers have asserted that it is possible to spray Roundup over the top of hazelnuts, without damaging them, at certain stages of growth. We will test the susceptibility of hazelnuts to Roundup and other herbicides in 2013, using “throw-away” plants we transplanted to the field in St. Paul in fall 2012.

Outreach.
Our third Midwest Hazelnut Growers Conference was held March 2-3 in Decorah Iowa with the proceedings posted on the Midwest Hazelnut website www.midwesthazelnuts.org. In addition, project personnel (Lois Braun) presented at Norm Erickson’s annual hazelnut field day on October 1st 2011, and will do so again on October 6th 2012, and hosted a field day at the hazelnut research planting at Rosemount on July 26, 2011. She also gave tours of her research plots at the Erickson Farm to people who attended the Slow Foods Minnesota Annual Event there on May 20, 2012.

Objective 4. Increase Grower Knowledge.
• We are getting ready for our 4th Annual Upper Midwest Hazelnut Initiative Conference, to be held March 1-2, 2013 in Eau Claire, WI. Attendance at the 2nd and 3rd conferences, in South St. Paul, MN and Decorah, IA was 48 and 57 respectively. Attendance at these conferences was 48 and 57 respectively. Attendees were a mix of current and prospective growers, and a few who were merely curious. Research findings from this project were reported at both.
• May 20, 2012— Lois Braun gave field tours of the hazelnut germplasm performance trial at Hazelnut Valley Farm, Lake City, MN, which was hosting the Slow Foods Minnesota Annual Cook-out, with over 100 attendees.
• July 26, 2012—Lois Braun hosted a tour of the hazelnut research at the Rosemount Research and Outreach Center, where at least six different research trials are being conducted, as part of the Rural Advantage “Walk-n-Talk” series, with five attendees.
• October 6, 2012—Annual meeting and field day of the Minnesota Hazelnut Foundation at the Hazelnut Valley Farm, Lake City MN, an annual event comprised of seminars in the morning and outdoor sessions in the afternoon, with about 50 attendees.
• Finally, the Upper Midwest Hazelnut Website (www.midwesthazelnuts.org) has proven to be an extremely useful channel to get research results out to growers just as soon as it is developed. The site typically has about 200 visitors per week.

Impacts and Contributions/Outcomes

Impact and Contributions/Outcomes
The benefits of this research to growers and consumers will mostly be realized when new germplasm is available from this work. In the meantime, growers are already benefitting from preliminary research results on best management practices, as we share them at conferences, field days, and via the Upper Midwest Hazelnut Website.

Collaborators: