The New Zealand Walnut Growers Manual

© 2003, 2005 and 2020, New Zealand Walnut Industry Group

All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording or by any information retrieval or storage system without prior permission in writing from the publisher.


Jenny Lawrence

Keith McIntosh

Keith Young

David Murdoch

David Ryde

Tony Davoren

David McNeil

Heather North

Clive Marsh

Ralph Brown


Ross Jamieson

Jenny Lawrence

Valda Muller

Otto Muller

David Murdoch

Blair Miller

David McNeil


Lucy Gardner

Janaki Kandula

Linda Gardner


The Production Team

Ralph Brown (Chair)

Heather North

Mike Caldwell

Donna Hiser

Murray Dill

Web Site Edition

Nelson Hubber

Online edition

This is an online version of our Growers Manual which was published in 2005. It has not been updated since then. Orchard practice is still much the same and the manual provides a very sound basis for how to plan and run a walnut orchard and look after the trees. However the reader should be aware that some things evolve and change and more up to date information about some topics may be found in other parts of this web site and from fellow members. 

Your committee, with help from a range of experienced growers, have produced more than 60 pages of useful information.

Currently out of print. A complete printed and bound edition will now cost $65 plus $9.00 postage including GST (a total of $74.00 incl).  For delivery to a Rural Delivery address please add a further $2.00. [for postage in NZ only]  

Profits go to your Walnut Industry Group’s contribution to the research.

If you would like a printed version of this manual please order it through the “Contact Us” form on this web site.

About your manual

Think of it as a living document.

Your manual draws on the expertise of many of your industry’s most knowledgeable people and represents more than 20 years of research and growing experience. As knowledge of walnut growing in New Zealand changes, so can your manual. 

The manual is physical evidence of the co-operative nature of your industry. For many years, growers, propagators, scientists, processors and technical experts have openly shared their knowledge in the interests of developing a thriving walnut growing community and a world-class product. Please give feedback about your experiences Sharing of ideas keeps our community alive.

The manual is as close as we can get to a consensus. As you would expect, some of our contributors have differing views on particular subjects or use alternative techniques in different regions or soil conditions. The differences are minor and where it’s useful, we have acknowledged them by suggesting more than one option.

Each chapter stands alone, with its own page numbering. You’ll find that there’s very little cross-referencing but the indexes will make it easy to find more detail in another chapter or section.


1.1 Introduction

Welcome to one of New Zealand’s newest horticultural industries.

We are producing a healthy whole food in a sustainable way from trees and our growers are environmentally conscious.

The walnut varieties you’ll grow are the product of more than 15 years of research. They have been selected from 800 samples collected around New Zealand and chosen for their vigour, yield, and the taste and processing qualities of their nuts.

The demand for New Zealand grown walnuts is greater than the supply, yet New Zealanders are still only modest consumers compared with Europeans and Americans.  On average, we eat just eight kernels a year!  Many New Zealanders have never tasted their own country’s walnuts, so there’s potential to grow the market through education.  There’s market demand for certified organically grown walnuts and they command a premium.

There’s also a large import substitution market.  New Zealand imports 300 tonnes of walnut kernels each year and in 2002 we produced just 30 tonnes.  We estimate that approximately a third of the demand is from consumers for whom quality is paramount.  In this market we are confident that we can continue to provide a product that will appeal to New Zealanders.

At present, our customers are specialist gourmet food stores, health food shops, fresh fruit and vegetable retailers, hotels, restaurants, caterers, food processors and selected supermarkets.

Not all the product is sold as food.  New Zealand walnuts are also used for woodworking oil, in pharmaceuticals and there’s even a market for the shells in aviation – they are crushed and used to clean aircraft engines.

There’s potential to export, but we believe that we could only succeed with high value niche products.

Our best estimate is that demand in New Zealand for locally grown walnuts will remain strong although our product does have to compete with imported walnuts.

1.2 Choose your site carefully

Walnuts are surprisingly tolerant of different climates, but there are limits for commercial production. We often hear of people who have, or remember, trees in seemingly unsuitable places that produced walnuts, but it’s best to grow them in free-draining soils in the east of the North or South Islands.

The relatively dry climate in the east inhibits the development of walnut blight, but even there, blight is a significant management issue. (Our researchers are working on it.) The New Zealand climate is cooler than the optimum for the high growth rates that Californian growers see, though depending on your soil and irrigation, your trees could grow about metre per year. There may be an advantage in the cooler climate: we believe that it produces a better-tasting walnut.

Walnuts are reasonably drought hardy but for commercial production you will probably need a water supply for irrigation or be in an area with a reliable rainfall of between 700 and 800 mm. A dry climate with irrigation is best. When you are buying your walnut property you will need to consider how much water is available for irrigation and the consistency of the supply year-round – including any regulatory restrictions.

The location of your water is important too. For example, it is more economical to have a spring-fed supply, gravity fed from a hill, than to be faced with drilling a bore and pumping the water from 50 metres below the ground.  However, a deep well accessing aquifers or ground water is likely to be a more consistent supply and probably less affected by drought.

Some shallow ground water and dams have iron, algae, bacteria or sediment and you may need a filtration system.

If there is already a water supply on the property, it is worth making a few simple calculations to check that there is sufficient water. You should speak to your local regional council about water sources, water availability and restrictions.  If you are installing a new bore or other water abstraction system, you will probably need a resource consent.

Walnuts are very sensitive to inadequate oxygen in the soil. They can die quickly if their roots sit in stagnant water, and root rot can occur in over-wet soils. Check the soil maps for your site and dig holes to check the depth of topsoil, the size of the gravels or stones and for any clay pans that would hold water around the roots. You may be able to rip a clay pan with a winged implement, mole plough or backhoe to improve the drainage. Enquire about the height of the water table, especially during the growing season. Walnuts are deep rooting, so if it’s within three metres, it may stunt the trees’ growth.

A fertile soil does make a difference to the growth rate but it’s not as important as drainage and fertile soils can be significantly more expensive. Some growers prefer to put their resources into building up the fertility and structure of a lighter soil. A pH of 6 to 6.5 is ideal.  

You will also need effective shelter. If you don’t have it, you are not ready to plant walnut trees. Strong winds will restrict their growth and damage them which may make them more prone to walnut blight. Any advantage you might have in planting early will be lost.

Walnut trees tolerate cold temperatures – down to at least minus 10 degrees Celsius during winter. However, spring frosts will damage their shoots, leaves, catkins and flowers. You will need late leafing varieties in a frosty climate and may need to cover your young trees to protect them from frost.  They need about 800 hours below 10 degrees before they break out of their dormancy.

1.3 A brief look at the economics

Walnuts are slow to produce an economic crop. Your returns are likely to be negligible for the first seven years but will continue to develop as the trees close the canopy of your orchard.

Your production will vary depending on your management practices, soil type, shelter, irrigation and your luck with frosts. One of our reviewers believes that it’s reasonable to expect an average of two to two and a half tonnes per hectare once the trees are mature. An extended research project in at Lincoln University estimated that the best cultivars would be capable of yielding 3.5 tonnes of walnuts (in shell) per hectare at full production.  It is probably best to do your budgets on around 2 tonnes per hectare.

The price will depend on what you grow, the quality of your product and how you sell it. If you sell to a processor, you will receive more for walnuts grown on the grafted varieties chosen from the research than for wild or seedling walnuts. Organic grafted walnuts command a further premium, but you would need to have a certificate from a recognised authority to show that your property and practices comply with its standards.

The 2004 prices to growers from New Zealand’s largest processor were $10.50/kg for top grade grafted kernels and $6.50/kg for baking grade (seconds). Seedling or wild kernels will give you a return of $8.50/kg for top grade and $5.50/kg for seconds. (Note that kernel weight is around 40% of in-shell weight, depending on cultivar). Processors may choose to become approved to the quality standards administered by Nut Industries of New Zealand (NUTNZ).

Your setup costs will include the shelter and walnut trees and probably irrigation. They may include labour, consultancy fees and a tractor and mower.

Your ongoing costs are likely to include herbicide sprays (at least in the first few years), fuel for mowing grass, electricity for irrigation, accountancy and, eventually, labour and freight costs for harvesting.

It is an expensive, long-term investment, but compare it with an investment fund you would need to generate the same income. Just as important, growing walnuts is a very satisfying activity. For those who choose to play an active role in the walnut industry, it’s very interesting and sociable.

1.4 Your industry group

The New Zealand Walnut Industry Group is here to help you. NZWIG’s aims include encouraging discussion and learning with publications (including your manual) and field days. We also instigate research projects and work with the research team as they search for answers to key questions for growers.

So far, NZWIG has organised regular field days and social events, produced this manual and an information pack and organised a conference for the industry. It has also set up agreements with walnut propagators to ensure that growers have access to approved cultivars. NZWIG has commissioned research on managing walnut blight and is setting up trials on new cultivars and rootstocks as well as a pruning trial.

NZWIG is not just for growers. We welcome all stakeholders in the walnut industry and draw on their expertise.

In the next few years, The New Zealand Walnut Industry Group will focus on these goals:

  • To establish and promote quality standards for propagation, growing, processing and sales and marketing.
  • To achieve and maintain a stable industry through collaboration between propagators, growers processors and retailers.
  • To provide information, including benchmarking, education and social contact to enhance our members’ skills.
  • To create a high market awareness that New Zealand-grown walnuts are a superior product.
  • To increase our knowledge of propagation, growing, harvesting, post-harvest handling and processing.


1.5 NZWIG Propagators’ scheme

The NZ Walnut Industry Group operates a propagators’ scheme (which evolved from a scheme initiated in 1985 by the original Walnut Action Group of the NZ Tree Crops Association).  The scheme serves three purposes:

  • To provide information on the most promising cultivars to both growers and propagators.
  • To provide funds for research into existing and new cultivars.
  • To pay royalties to the owners of the original parent trees.

NZWIG encourages you to purchase your trees from one of the participating propagators.  For their names and addresses, please contact the administrator of the propagators’ scheme, Donna Hiser, using the address details on the title page of the manual.

The propagators will provide you with a certificate confirming the cultivar identity of your trees.  You’ll find the certificate a useful way to confirm to processors of your nuts, or future purchasers of your property, that your trees are what you say they are.


Reproductive Life of a Walnut Tree
How buds form on walnut trees
Pictures of walnut diseases



Let’s assume that you have carefully considered the soil types and climatic conditions, purchased your property and you are ready for the major undertaking of developing a walnut orchard.  It’s well worth setting aside the time you need to plan an efficient and economic walnut orchard and for avoid costly mistakes and unnecessary work.  Your planning should include:

  • block sizes
  • shelter from wind
  • walnut planting distances
  • selection of walnut cultivars
  • orchard floor preparation

Your decisions on the layout are also vital for designing your irrigation system.

CHAPTER 1 – Shelter

If you are considering planting a walnut orchard in New Zealand and not planting shelter – forget it!  Walnut trees, especially young ones, are very susceptible to wind damage and even large trees can suffer broken limbs.

1.1 Block Sizes and Shelter Requirements

The appropriate size of your blocks will depend primarily on the height of the shelterbelts.  The standard guide works for shelterbelts more than 12 metres high: expect good shelter for a horizontal distance of about eight to ten times the height of the shelterbelt. (see Figure 1.1).

Effect of Shelter on windspeed
Figure 1.1: Effect on wind speed of a shelternelt graphed at various distances from the shelter belt. (The sheltewrbelt is shown as the thick black line. The distance of benefit is shown in multiples of tree heights. For example, at a distance from a medium density shelter belt of 10 times the tree height, the wind speed is onoy 60% of the full speed

Some growers buy land with good shelter already in place, but most establish new shelterbelts.  Fast shelter species grow at a rate of one and a half to two metres per year in good locations, but a water-loving shelter species grown on a bony site in the full force of the wind may grow as little as half a metre per year, even with irrigation. If you hope to plant walnuts two years after establishing shelterbelts, the blocks should be around 50 m x 50 m or 60 m x 60 m. If you prefer your blocks to be 80 metres across, you must be comfortable with waiting three or four years before planting your walnuts. Many experienced growers believe it’s worth waiting because the larger blocks will make it much easier to use machinery in your orchard.

A good option is to make use of temporary shelterbelts.  Allow the temporary shelterbelts to grow for say six to eight years to give you early shelter for small block sizes, and then rip them out when the permanent shelterbelts reach sufficient height.

As well as block size, you’ll need to consider some other issues:

  • Presence of existing shelterbelts on neighbouring properties – can you depend on these shelterbelts in the long term, or is it wiser to plant a parallel shelterbelt on your own land?It is very difficult to establish young shelterbelt trees right next to a mature pine or eucalyptus shelterbelt unless you create a deep ripping line along the boundary.
  • Selection of shelter species according to growth rate, final height, suitability for site, and the time the deciduous trees shed their leaves.
  • Your soil type.
  • The dimensions and topography of your land.

Horticultural advisers and tree nursery staff may be able to provide specific, on-site advice, to help you assess your situation and requirements and weigh up the options.

1.2 Shelter Species

Most growers, starting from a bare block, aim to obtain the tallest possible shelter in the shortest possible time.

Crows Nest Poplar is a modern, rust-free version of the Lombardy poplar and has few rivals for speed of growth and final height. It will grow to 35 metres – if it receives sufficient water.  On a damp site Crows Nest Poplars can grow two metres per year, but on a drier site in the full force of the wind, one metre per year is more realistic.  Many growers use them as shelterbelts throughout the orchard.  They are often used at least on the perimeter of the property, with slower-growing species for internal shelterbelts.  The main drawback is that poplar roots tend to spread out many metres into the orchard, sucking water and nutrients from the cropping trees.  When that happens, orchardists tend to rip the roots along the poplar line every couple of years but it can be an expensive process.

You will probably want deciduous shelter trees such as poplars on the north side.  They provide shelter during the late spring, summer, and early autumn, when the walnuts are in leaf, but do not cast a cold frost shadow in early spring when walnut trees are budding out.  You can also use other species, such as Tasman Poplar which has a final height of 30 metres and a growth rate of up to one and a half metres per year on a damp site.  Tasman Poplars have a slightly more bushy form than the Crows Nest, so need more side-trimming – which is also expensive.

Italian alder (Alnus cordata) is increasingly planted as a shelter species for walnuts.  It is deciduous, but holds its leaves until late in the autumn so they are not on the ground during walnut harvest.  The roots go deep into the soil rather than spreading out into the orchard area.  They grow quickly in warmer, damper sites (one and a half metres per year), and rather more slowly (though still one metre per year) in drier, cooler sites.  They can grow to 30 metres, but 20 metres is more common on lighter land.  Twenty metres is probably minimal, in the long term, for sheltering walnut trees but with their other qualities it’s well worth considering Italian Alders for at least the internal shelter.  Growers in Central Otago report that Italian Alders are vulnerable to the heavy frosts there.

Willow (Salix matsudana 1184) used to be widely recommended as a deciduous shelter tree because it grows fast, tall, and has a naturally narrow form with a final height of 25 metres and a growth rate of one and a half metres per year on a damp site.  It is, however, best suited to those damp sites, and requires careful and frequent irrigation if it is to flourish in areas suited to walnut growing.  You would need to trim the crown each year and probably rip the roots regularly so it’s an expensive option.  Willow does have some useful characteristics.  Its new leaves will give you early spring shelter.  The willow doesn’t produce suckers so you can remove it in one operation.

Evergreen trees are valuable on the perimeter of the property to block cold south winds.  Each region of New Zealand has characteristic cold winds from certain directions at certain times of the year.  For example, the prevailing wind from spring to autumn in many parts of Mid to North Canterbury is the cool north-east wind, so evergreen shelter to the east may also be valuable.  Certain Eucalyptus species, macrocarpa (final height 25 m, growth rate one and a half metres per year) and other Cupresses species all grow fast, and are often used as perimeter shelterbelts to the south.  In particular, Eucalyptus nitens (final height 50 m, growth rate two metres per year) is frost-hardy, and grows much faster than poplars on a hard site and in the full force of the wind.  The nitens species is not drought tolerant, so you would need irrigation.  Some growers use Pinus radiata(final height 25 m, growth rate one and a half metres per year) but it can be very expensive to maintain as an effective shelter.

There are some shrubs and small trees you could use to fill in the base of a shelterbelt including: ovens wattles, lacebark (Hoheria populnea) broadleaf/kapuka (Griselinia littoralis), Pittosporum tenuifolium, Karamu (Coprosma robusta) and many others.  The selection will depend on your climate.  Your nursery or regional council should be able to provide you with information about eco-sourced native trees.  In Canterbury, ask Environment Canterbury for the publication, ‘Establishing Shelter in Canterbury with Nature Conservation in Mind’.

In summary, the factors to consider when selecting shelter species include:

  • The rate of growth and final height of the trees
  • Climatic, soil and moisture requirements
  • The tree’s natural shape and porosity; side-pruning requirements and the cost of the pruning
  • The cost of the trees
  • The penetration of the shelter trees’ roots into the walnut growing area and the cost of ripping the roots every couple of years
  • The effect of autumn leaves on your harvesting.

Note that all of the trees we’ve suggested need irrigation on most sites suitable for walnut growing.  Also, it’s best not to top shelterbelts for walnut orchards because you’ll need them tall for such a large crop tree.

Good wind porosity (the ability of the wind to pass through the shelter line) of 40%-50% is important, as shown in Figure 1.1.  A porous shelterbelt will slow the wind velocity, but still allow air to pass through.  A solid shelterbelt, on the other hand, will block the wind, which will then be forced up and over the top and then dumped into the low pressure zone on the leeward side.  The effects can be very damaging.  Choose a suitable species and use side trimming and other pruning methods to give your shelterbelts the porosity you need.

1.3 Planting of shelter trees


If you are planning any whole-of-property operations such as cultivation, levelling, planting of orchard floor grass, fertilisation, etc., it is more efficient to do them before you split the property into small blocks with shelterbelts.

Pegging Out

Mark out your shelter lines by pegging out the intervals.  They can be measured out along perimeter fences using a long tape measure or light wire rope with the intervals marked on it.  Spray the pegs with dazzle paint so that you can see them in the distance.

Shelter layout

Plan to plant at least five metres away from overhead power lines otherwise a council operator is likely to come along and give your trees a prune!  There is little point in planting Crows Nest Poplars directly under power wires and then having to top them regularly.

If your shelter trees require side pruning, allow space for a machine to do the job. Also allow space for any root ripping and for heavy machinery including tractor mowers, possibly fertiliser or spray trucks, and mechanical harvesting equipment.  You will probably need to allow five to six metres.  Create headlands eight to ten metres wide so that machinery can turn at the ends of rows.  Make your gateways through the shelterbelts at least four metres wide.


After pegging out, rip the shelter lines with a deep winged ripper.  It will make the planting very much easier.  If the soil is compacted, ripping will loosen and aerate the soil, and allow water to move easily into the rip line.  It will also help the roots to penetrate in the first few years.  Winged rippers are available from many tree nurseries on a daily or hourly rate and can be attached to the three‑point linkage on tractors over 70 hp.


The spacing of your shelter trees will depend on the species you choose.  You can plant Crows Nest Poplar, with its narrow form, closer together than Eucalyptus nitens.  There is a trade-off between the close spacing that seems ideal early on when the trees are small (say, one metre between trees) and the two to three metres spacing that is more appropriate when the trees are mature.  Some growers initially use a close spacing of one metre or less for poplar and then thin out every second tree.  Alternatively, you can leave the trees at their close spacing but side-prune more frequently to maintain the porosity.  Other growers plant poplars at their final spacing of two to four metres, but place a bushy species between each one for early shelter.  You can use ovens wattle, Pittosporum tenuifolium, Coprosma robusta, and many others for the purpose and remove them later if you need to.

You could also alternate Crows Nest Poplar with the slower-growing Italian Alder, then remove the poplars once the alders are big enough to provide adequate shelter.


Plant the deciduous species such as poplar and alder between June and August and the evergreens in spring.

If you have a large block of land, and wish to plant all your shelter in one season, try to enlist help from help friends and family.  (Your city relations may be happy to cancel their gym membership during planting season.)  As a very rough guide, you may be looking at 2,000 shelter trees to be planted on a 10 hectare property (depending on your block size, tree spacing, etc.).  Two people could expect to plant about 200 trees into a rip line during the course of an 8-hour day, depending on the soil – and on your stamina.  Planting is very much slower if you are digging into soil that has not been ripped.

It is important to plant poplar deep into the soil because they will form roots all the way up the section of stem in the ground.  They will then be more robust in wind and more able to harvest water and nutrients.  Willow and poplar cuttings should be at least one metre long and a thumb thickness in diameter.  If the cuttings are already rooted, you will need a wider hole and they should be planted between 200 and 300 mm deep.

Use a board or bamboo pole with markings at your chosen tree spacing to save you time measuring out your tree spacing.

1.4 Managing your shelter trees

To promote good growth of shelter trees in their first and subsequent years:

  • Reduce the competition from weeds Most growers spray with herbicides such as glyphosate (the generic form of Roundup), Buster, or a pre-emergent spray like Simazine to kill grass and other weeds.  It is best to spray before bud burst in the spring so that spray cannot fall on tree leaves.  A pre-emergent spray will probably last all season (because it is somewhat residual in the soil) whereas if you use a knock-down spray, you will most probably need at least one more spray during the season.  Mulch can also suppress weed growth, as can cultivation during the growing season, though it is a more labour intensive option.
  • Fertiliser Apply a general fertiliser around each tree before September and another during October or November.  Poplars and eucalypts are gross feeders and will make use of extra nutrients even if there are no apparent deficiencies in your soil’s nutrients.
  • Irrigation It’s common practice to run lateral tubing along all shelterbelts and install an irrigation dripper beside each tree.  Your tube diameters and dripper flowrates should be selected as part of your overall irrigation system design.  Pressure-compensating drippers are needed if the water pressure varies throughout your irrigation network, but they may block with dust and insects more easily than non-compensating drippers.  ‘Holey pipe’ is also an option.  Drippers of flowrate 8 litres/hour are commonly used but other flowrates are available.
  • Tree protection – shelter boxes and spray guards The dangers to shelter trees are hares/rabbits, wind and spray drift.  When you are considering the economics of protecting your shelter trees, take into account not only the cost of the shelter tree, but lost growing time.  If the growth of your shelterbelt is held back for a year, your walnut planting date will also be delayed by a year, and then your first walnut harvesting date will be delayed by a year.  Shelter boxes, which can be 750mm high and are held up by a wooden stake, provide excellent protection from all the dangers, but they cost more than the shelter trees they protect.If you decide to use shelter boxes for your walnut trees, buy them early and put them around the most vulnerable of the shelter trees in the meantime.  Otherwise, it is a good idea to put a plastic bag or sleeve around the bottom 300mm of the trees. You will deter browsing hares and rabbits and protect the young tree from spray drift during the growing season.  Tripel paint or egg mixture can help to repel the hares.
  • Farm animals: Grazing animals with small trees can easily go wrong – stories abound of whole blocks of trees being eaten by sheep, goats, horses or cattle.  Unless you are prepared to fence off all shelter lines or move electric fences continuously, keeping farm animals with trees is not worth considering.

As the shelterbelt ages, you will have to consider:

  • Side trimming It is necessary to side-trim most species every two or three years, apart from very narrow-crowned trees like Crows Nest Poplar, planted at two to three metre spacing.  Side-trimming has several purposes: preventing shelter trees from branching out into orchard areas, maintaining the porosity, and maintaining a fine twiggy branch structure right to the base of the trees which would otherwise open out into large branches and few leaves, in most species.  Topping is generally not recommended, unless you have very small blocks with very tall shelter, but very few growers have that problem.
  • Possible thinning Once shelterbelts become tall and thick, it is possible for frost to ‘pond’ behind them and if they become too effective at stopping air movement, they may promote humid conditions that are suitable for blight development.  You could thin out every second tree (or similar) to regain suitable porosity.  If your shelterbelts become tall enough to shelter two blocks, you could remove whole lines of trees, leaving larger blocks for easier management.
  • Root ripping If you have chosen poplars, willows or eucalypts for your shelterbelts, keep in mind that within a few years you may need to rip the roots to stop them stealing water from your walnuts.
  • Waiting Some growers are able to harvest hay or silage while they wait for the shelter to grow, though most contractors do not like working in small orchard blocks, so you may have difficulty in finding someone to cut and bale it.  Even if you do, you are unlikely to receive a high price.  Another option is to lease out the blocks to someone else for two years, making sure that your irrigation lines and shelter trees are well protected during that period.  In practice, most growers simply mow the grass and leave the clippings to break down and return to the soil.

CHAPTER 2 – Setting up your orchard

This section lists four tasks that are generally best carried out before you plant your shelterbelts, or at least before you plant your walnut trees.  Some of the tasks are covered in more detail elsewhere in this manual.

2.1 Soil analysis

Growers use soil analysis to identify and address any gross nutrient deficiencies and then fine-tune the nutrients on an ongoing basis from the results of leaf-testing the walnuts.  It’s easier to broadcast fertiliser to address any major imbalances before you split the property into small blocks, so it is a good idea to check the nutrient status of your soil before you plant your trees.  You’ll find detailed notes on soil and foliage sampling and analysis, and interpretation of the results, elsewhere in this manual.

You should find out the names and characteristics of the soil types on the property before you purchase.  It’s valuable information, because different soil types will often have different nutrient status, and can have very different responses to the water you apply.  For example, a soil with low organic content will have less capacity for holding nutrients, and fertilisers will be more easily leached through into the sub-soil or groundwater.  A light, shallow soil will have less water-holding capacity, and thus will need to be irrigated in smaller, but more frequent, doses than will a heavier soil.

2.2 Installing irrigation

Designing and installing your irrigation system is a major undertaking, and many growers choose to have professional irrigation engineers do it for them.  However, it is very useful to understand the basics, and they are covered in the chapter Designing and Installing Your Irrigation. The New Zealand Irrigation Manual is another valuable resource and it’s available at some libraries. You can find out more at Environment Canterbury’s website:

With those sources, even a basic knowledge of engineering calculations will be enough for you to at least determine the total volume and flow rate of water you need for your property.  It’s best to install a bore (or other water source), pump and mainline at an early stage as it will be expensive and disruptive to install them later.

Before beginning the design for layout and sizing of sub-mains, laterals and emitters, you will need to make decisions about the size of your blocks and the spacing of your shelter and walnut trees.  Those decisions will allow you to work out the total number of trees to be irrigated by dripper (usually the shelter trees) and by micro-sprinkler (usually the walnut trees).  Separate controllers must be installed for areas serviced by drippers and micro-sprinklers.

Shelter trees are usually planted in the winter and early spring and you won’t need to irrigate them till late spring or early summer.  The lateral tubes and drippers for irrigation of shelterbelts are usually laid after planting of the shelterbelts but before the soil dries out. Lateral tubing for irrigation of walnut trees is usually mole-ploughed to a depth of about 100 – 200 mm, and micro-sprinklers plugged into it while the soil in the rip line is still soft.

Since most growers plant walnut trees in late spring to early summer, you should install your irrigation before the walnuts are planted out so that the trees can be irrigated as soon as they are in the soil.  Make sure you place the irrigation tubing well clear of the future location of your trees.

2.3 Levelling and cultivation

Most growers use micro-sprinklers now instead of flood irrigation so levelling the ground is not as important.  However, you will need to allow room for tractors and other machinery to move around easily in the orchard, and the ground surface should be smooth enough for you to harvest the walnuts efficiently.  Some growers cultivate the soil before planting to help the trees develop healthy roots, but working dry soil in summer can damage the soil structure. Late spring or autumn is usually a good time.  If you are planning to use a tree auger or back-hoe you probably won’t need to cultivate the whole block.  (If you are using an auger in fine soils, be sure to break up any glaze that forms in the hole.)  Avoid compacting wet soil by driving over it with heavy machinery.

2.4 Orchard floor planting

Some growers plant their block in a slow-growing orchard grass.  This would reduce the time you spend mowing, but the cost of the seed may well out-weigh the benefits.  Slow-growing grass is seldom used in large orchards.  Another option is to plant a species that makes good hay, such as clover.  The value of the hay may help you convince a farmer or contractor that it is worth the effort of mowing your small orchard blocks.  The majority of growers, however, simply leave the grass in their purchased property as-is, and mow as required.

One option is to mow infrequently, but spray out the whole block every few years with a glyphosate spray that kills grass and many other weeds, but only knocks the clover back.  In this way, you encourage clover growth and keep control of grass growth.  Clover ‘fixes’ nitrogen from the air and adds it as a natural fertiliser into the soil.

It used to be common practice to cultivate soil during the growing season to eliminate competition from grass and weeds and to improve the flow of flood irrigation before micro-sprinklers became common.  Few growers cultivate now because frequent cultivation damages the structure of the soil and tree surface roots.  Exposed bare soil is vulnerable to wind erosion and to compaction from rainfall and irrigation.  The lack of vegetation reduces the addition of organic matter from plant leaves and roots, so there will be a poorer habitat for soil micro-organisms, worms and insects.  Even so, cultivation may be an option if you have limited irrigation, because you will reduce the competition from weeds.

If you are growing organically, use the future orchard floor as a source of mulch.  You can exploit clover’s nitrogen-fixing qualities, and deep rooting plants such as chicory and others can make nutrients available to your trees.

You’ll find information about managing the orchard floor in Section 3.

CHAPTER 3 – Choosing Your Cultivars

3.1 New Zealand selected walnut cultivars

In the early 1980s, members of the South Island Walnut Action Group (a branch of New Zealand Tree Crops Association) began working with researchers from Lincoln University in Canterbury to gather promising walnut cultivars from throughout New Zealand.  The cultivars were assessed for nut quality (taste, colour), suitability for commercial growing and processing (size, shape, shell seal, storage properties), and ‘crack-out percentage’ (how full of kernel the shell is).  The best cultivars were then grafted and grown on at Lincoln.  Over the following years, the trees were assessed for yield, biennial bearing, form, flowering times, and other characteristics.  By the early 1990s, the researchers were able to recommend four New Zealand selection walnut cultivars and orchard management practices covering tree spacing, pruning and suitable pollinators.  Many of our recommendations in this manual are based on the results of those trials, augmented by more recent experience in the growing walnut industry.

New Zealand is very unlikely to be able to compete internationally on price, so high product quality and niche products are more appropriate market strategies for the walnut industry than volume and low cost.  Several of the selected cultivars have excellent quality characteristics, and we owe a great deal to the foresight of the walnut industry pioneers.  At least one walnut processor is beginning to market varietal walnut products, and walnuts from grafted New Zealand selections attract a higher price to the grower from several processors than do seedling walnuts of unknown genetics.

A walnut planted in the ground will grow into a seedling whose nut characteristics only appear when it begins bearing.  Therefore, most growers plant grafted trees – where scion wood of known parentage has been grafted onto a rootstock.  The fruiting wood on this tree is then essentially a clone of tree from which the scion wood was taken, and the nuts will have exactly the same characteristics.

The propagators participating in the NZWIG propagators’ scheme have agreed to a set of colour codes for newly-grafted walnut trees. The codes will help you to identify the selected varieties too. Rex-blue, Meyric-red, Ble 300-yellow, G120-white, G139-pink, Tehama-green.

3.2 Crop trees

The New Zealand Walnut Industry Group recommends a range of cultivars that performed well in the trials at Lincoln University, particularly the cultivars Rex and Meyric.  So far, the selections have been grown mainly in Canterbury conditions, but Rex, Meyric and Stan are also doing well in Central Otago.  The North Island Walnut Action Group is running a trial in the Wairarapa and the New Zealand selections are looking very promising.  Several of the European and American varieties perform successfully in Central Otago (Vina is particularly favoured) and Marlborough (where at least one grower prefers Serr).

Rex is the best-performing cultivar overall in the New Zealand selection.  It is also known by its trial number,152.  Rex produces a crop early in its development.  It is a smaller tree than many but produces nuts on its lateral shoots[1].  The nuts have a medium to high crack-out percentage (the percentage of kernel to total nut weight) of 40 to 45 per cent.  The Rex kernel is blonde and sweet.  It has a significant health advantage because its fatty acid profile makes it the most efficient nut of the New Zealand selections for reducing blood cholesterol levels.  One processor is already marketing Rex as ‘the health nut’.  Rex appears to be more resistant to walnut blight than any of the other cultivars in the Lincoln trial which means that you may not need to spray as often.

Meyric (trial number 1199-4) is also a very good cultivar, though it is terminal rather than lateral-bearing, so is slower to produce in its early years.  It also appears more susceptible to damage from blight than Rex (though still less so than many other cultivars).  It has an extremely high crackout percentage at 47 to 52 per cent.  The kernel is a honey blonde.

Meyric and Rex kernels maintain their colour and flavour for up to 15 months.

The other two main New Zealand selections were Dublin’s Glory (trial number 143) and Stan (Ble 300), though they are now less strongly recommended.  The nuts from both are of a high quality but Dublin’s Glory leafs out too early for frost-prone areas of Canterbury and Central Otago. There is some evidence that Dublin’s Glory is more prone to blight than Rex or Meyric. Stan produces kernels which are very light in colour and there is some resistance from consumers because they believe that they are not ripe.

Other cultivars that gained high scores in the original trials are 151, NZ Purple (G139), Vina, Tehama and Esterhazy.  Vina is planted extensively in Central Otago, where it performs well, preferring the hot summers.  It has a high yield.  It tends to be susceptible to blight, but this is less of an issue in the dry Central Otago conditions.  In Canterbury orchards the cultivars other than Rex and Meyric are used mainly as pollinators.  It is certainly worth considering them, in the light of your specific climatic conditions, but it can be difficult to obtain large numbers of cultivars other than Rex and Meyric, unless you graft them yourself or order well in advance.

The New Zealand Walnut Industry Group has initiated a new search for walnut cultivars. Though Rex and Meyric are excellent, NZWIG thought it wise to keep up the forward momentum with walnut genetic material and to widen the available choices of good cultivars.  The nut quality would have to be at least a match for Rex and Meyric for a cultivar to be entered into any new trial.

3.3 Pollinators

Pollination in the orchard

Viable pollen must be available at the same time the female flowers are receptive. Generally pollination is not a problem with walnuts other than very late flowering varieties such as Franquette.  Walnut pollen can travel for many kilometres, so even apparently isolated trees usually get pollinated.  Walnuts are, in fact, self-fertile, but the chances of self-fertilisation are not high. The shedding of the pollen and the receptivity of the female flower do not overlap very much within one cultivar, and pollen is viable only for a short period.  Very warm weather causes the male flowers to release the pollen early, and rain, strong wind, or low temperatures can reduce the amount of pollen dispersed. Such conditions make relying on self-fertilisation more hazardous.

Rex and Meyric are able to pollinate one another.  Rex’s mid pistillate bloom coincides with the time Meyric sheds its pollen but the overlap is only a few days and it’s not the best pollinator for Rex.  It’s likely that simply planting them both as crop trees would be sufficient for pollination but some experienced growers believe that it makes sense to include about five per cent of other varieties as pollinators and up to 20 per cent if you are planting late-flowering varieties of crop trees.

When planting pollinators you’ll need to take into account some other significant factors.

Wind direction:  For maximum benefit, you should note the direction of the prevailing wind and plant your pollinator trees upwind of the main crop trees.

Harvest:  If you plant the pollinator trees in solid rows in a common direction it will be easier to keep their crop separate at harvest time.  The nuts may fall at a slightly different time, which can also help.

Tree removal:  If you are using temporary trees in high density spacings, plant your pollinator trees in the rows you intend to be permanent.

Selection:  Pollinators such as Rex, Meyric, Stan (Ble 300), G120 and Tehama also scored highly in the Lincoln selection trials as cropping walnuts.

[1] Some varieties can bear fruit on shoots sprouting from the full length of one-year-old wood; these are called lateral bearing varieties. Others can bear fruit only from a terminal shoot at the end of one-year-old wood, and these are called terminal bearing varieties.  Lateral bearing is a valuable characteristic, leading to higher yields in the early years.

3.4 Compare the cultivars

This section contains tables and graphs of information to help you compare the characteristics and performance of walnut cultivars, match crop trees with pollinators and predict when your cultivars will burst their buds in spring. Bud burst is a critical time for spraying against walnut blight.  If you are in a frosty area, it is worth considering the later leafing varieties as they have lower likelihood of damage from spring frosts (this is mainly an issue in the early years when the trees are small).

How to use the next table

We’ve adopted Rex, the current most popular New Zealand cultivar, as our standard for comparing the leafing and flowering dates. (Until now we have used the American standard, which is the cultivar Payne, but hardly anyone grows Payne in New Zealand.)

The third column of the table records ‘Start of bud burst’ for Rex as 0  (which is October 2 in Canterbury) and Meyric as    -2; so if you have Meyric, you should expect bud burst for those trees to occur around September 30 (i.e. two days before Rex). Bear in mind that bud burst is a gradual process and the actual timing will depend on the climate in your area and the weather in early spring.

The figures in brackets near the bottom of column 3 are the equivalent bud burst figures for the same cultivars grown California. We’ve added them because they reveal interesting information about New Zealand walnuts. (The figures are all converted to the Rex standard.) The figures suggest that bud burst (and also female flowering and male flowering) all occur sooner in spring in California than in NZ. Our slower response to spring is probably explained by our cooler climate, particularly during early spring. Notice too, in columns 6 and 7, that the kernels have significantly greater mass and that the crackout percentage is higher in California. Our walnuts are smaller and lighter but, we believe, significantly better tasting.

Column 8 tells you whether the cultivars are lateral or terminal-bearing. A terminal-bearing cultivar produces fruit only near the ends of its one-year-old branches. A lateral-bearing cultivar produces fruit on spurs along the full length of its one-year-old branches. It helps to explain why a lateral-bearing cultivar such as Rex produces more fruit than a terminal-bearing cultivar such as Meyric in the in the early years and sooner after hard pruning. The distinction between lateral and terminal bearing is not a big issue after the first few years of production.

After the table, we’ve also included a graph of cultivar flowering dates so that you can easily match crop trees with pollinators by looking for the overlaps between male and female flowering.  Note that the two most popular cultivars, Rex and Meyric, are able to pollinate each other to a large extent, though most growers also include other pollinators in their orchards.

The yield data in columns 9 and 10 is from the Lincoln University trial block.  The data summarised here was collected when the trees were 8–11 years old, so provides only an early estimate of yield, but is the best data we have at present.


3.5 Cultivar profiles (of named varieties)

The following pages provide a table of information for each of the main walnut cultivars.  Some of this information is also provided in the preceding pages, in the cultivar comparison graph and table.  Nut quality assessments were part of the original walnut search, and yield, pollination and vigour data was collected at the Lincoln University trial block.  This work was done by David McNeil and others at Lincoln University, and members of NZTCA Walnut Action Group (South Island).

Each profile contains the following information (where it is available):


Table item


Tree characteristics


The spacings at which you plant your trees may well be influenced by their vigour and final size.

Ultimate size

Size at maturity.


Vase-like forms are difficult to train to a central leader, whereas apically dominant forms more naturally take this shape.

Blight resistance

Observations from Canterbury.

Bud burst

Stated in number of days after Rex bud burst, which is approximately October 2 in Canterbury.

Peak pollination

The mid-point of male (catkin) flowering, stated in number of days after Rex bud burst.

Peak receptivity

The mid-point of female (pistillate) flowering, stated in number of days after Rex bud burst.

Bearing habit

Terminal (nuts are borne on spurs only near ends of one-year-old branches) or lateral (nuts are borne on spurs along full length of one-year-old branches).

Yield of 8–11 year trees (kg/tree)

An average of yields in years 8, 9, 10 and 11, (across 5 or 6 replicates) for trees at the Lincoln University trial block.

Estimated full-production yield (t/ha)

Yield in kg/tree converted to an estimated yield for the cultivar at full production (full canopy cover).  These early estimates from trial block are probably slightly conservative.



Table item


Nut characteristics

Size and shape

A larger nut with attractive shell is important for in-shell market but not for processing.


As rated by taste panels.


Poly-unsaturated fatty acid. A low level indicates stability in storage but the higher the level, the better for your health.


Testing at Lincoln University repeatedly showed consumer preference for light coloured kernels.


Weight of the kernel as a percentage of the total in-shell nut weight. It’s a very important characteristic. We have given a range where that’s possible, but otherwise an average. Irrigation is the most important factor in the variation between orchards. Most of the figures are from results at A Cracker of a Nut processing factory. (Previously many figures were based on small samples at Lincoln University.)

Rex (W152)

The most popular of the NZ selections by a considerable margin.  Lateral bearing nature gives high yield in early years and is relatively blight resistant (in Canterbury).  Nut has good taste and processes well.  The nut is quite small (probably too small for in-shell sales). The crackout is quite low, but the total yield from the tree more than compensates for it.  The tree is not apically dominant. It’s not easy to train but it’s important to give it a helping hand.

History: NZ-selection.  Fifth place in walnut competition and trialist in 1987 trial.  Original tree from Christchurch.

Tree characteristics



Ultimate size



Poor apical dominance

Blight resistance

Very good

Bud burst


Peak pollination


Peak receptivity


Bearing habit


Yield of 8–11 year trees (kg/tree)


Estimated full-production yield (t/ha)


Nut characteristics

Size and shape

Smooth, plump. Small round nut.


Mild, sweet.


76.5 (Highest equal for all tested. The other is Serr.)


Kernel pale, smooth and shiny.


45% initially, then 38-43% depending on irrigation.

Meyric (W1199-4)

The second most popular of the NZ selections.  Terminal bearing nature means yield in early years may be lower than lateral bearing types.  Medium blight resistance (in Canterbury).  Nut has good taste and processes well.  A reasonably large nut (large enough for in-shell sales) and the crackout is very high. Tree is vigourous and apically dominant and generally easy to train.

History: NZ-selection.  Trialist in 1985 trial.  Original tree from Hawkes Bay.

Tree characteristics



Ultimate size



Good apical dominance

Blight resistance


Bud burst


Peak pollination


Peak receptivity


Bearing habit


Yield of 8–11 year trees (kg/tree)


Estimated full-production yield (t/ha)


Nut characteristics

Size and shape

4.5g kernel. Plump, even halves.






Kernel pale and shiny.


Settles to 47-52% on maturity.

Stan (Ble 300)

Vigorous tree with good, apically dominant form.  Relatively large, high quality nut with a beautiful appearance. It is more prone to blight than some in our list.

History: NZ-selection.  Trialist in 1985 trial.  Original tree from Blenheim.

Tree characteristics


Very high

Ultimate size

Very large


Good apical dominance

Blight resistance


Bud burst


Peak pollination


Peak receptivity


Bearing habit


Yield of 8–11 year trees (kg/tree)


Estimated full-production yield (t/ha)


Nut characteristics

Size and shape

4.6g kernel. A long nut.






Kernel blonde



Dublin’s Glory (W143)

An early leafing variety, susceptible to late spring frosts in Canterbury.  Yield data from the Lincoln University trial was highly variable, probably due to it being caught by frost in some years.  A heavy and lateral bearer which (frost permitting) should yield very well in early years.

More susceptible to blight than most of the other varieties.

The low PUFA tells us that it is the most stable in storage and it’s a good candidate for in-shell sales.

History: NZ-selection.  Trialist in 1985 trial.  Carpathian seed.

Tree characteristics



Ultimate size



Poor apical dominance

Blight resistance


Bud burst


Peak pollination


Peak receptivity


Bearing habit


Yield of 8–11 year trees (kg/tree)


Estimated full-production yield (t/ha)


Nut characteristics

Size and shape

3.9g kernel. Large – though the kernel is not always plump.


Mild, sweet.




Shell bright


38-43% on maturity


A lateral bearer with good early yield, but considered difficult to train.  One Canterbury grower prefers this variety (though it looked less promising in the early years) – fifteen year old trees produced an average of 30 kg per tree, to give an equivalent yield of about three tonnes per hectare. The yield is consistent if the trees are not frosted.

It’s suitable for in-shell sales.

History: NZ-selection.  Third place in walnut competition and trialist in 1987 trial.  Original tree from Christchurch.

Tree characteristics



Ultimate size



Very poor apical dominance

Blight resistance

Opinions vary: between very good resistance and ‘more susceptible than most other selections’.

Bud burst


Peak pollination


Peak receptivity


Bearing habit


Yield of 8–11 year trees (kg/tree)


Estimated full-production yield (t/ha)


Nut characteristics

Size and shape

Plump.  Relatively large nut.


Mild, sweet.




Kernel pale




Low producer and prone to walnut blight, but the clean, bright, smooth shells make them option for in-shell sales.

W153 is frost-prone.

History: NZ-selection.  Sixth place in walnut competition and trialist in 1987 trial.  Original tree from Christchurch.

Tree characteristics



Ultimate size




Blight resistance


Bud burst


Peak pollination


Peak receptivity


Bearing habit


Yield of 8–11 year trees (kg/tree)


Estimated full-production yield (t/ha)


Nut characteristics

Size and shape



Mild, sweet.




Bright and clean.




A German cultivar, correctly known as G1239 but abbreviated to G139.  Most striking feature is that the kernel is purple/red in colour.  Possibly related (may even be the same as) a German nut called Red Danube.

G139 is prone to blight. The branches are brittle, so you would need good shelter.

History: German import.

Tree characteristics



Ultimate size



Good apical dominance

Blight resistance


Bud burst


Peak pollination


Peak receptivity


Bearing habit


Yield of 8–11 year trees (kg/tree)


Estimated full-production yield (t/ha)


Nut characteristics

Size and shape

3.5g kernel. Kernel is purple/red.


Sweet. Always near the top in taste tests.




Kernel pale and shiny.




Possibly the most promising of the golden-kernelled German cultivars.  A larger nut most suited for the in-shell (table) market.

History: A German import.

Tree characteristics



Ultimate size




Blight resistance


Bud burst


Peak pollination


Peak receptivity


Bearing habit


Yield of 8–11 year trees (kg/tree)


Estimated full-production yield (t/ha)


Nut characteristics

Size and shape

5.6g kernel.






Kernel golden.




A very late leafing and flowering cultivar, possibly suitable for areas where severe spring frosts are common, but would need to check on harvest date in case autumn frosts might be a problem.

History: A German import.

Tree characteristics



Ultimate size




Blight resistance


Bud burst


Peak pollination


Peak receptivity


Bearing habit


Yield of 8–11 year trees (kg/tree)


Estimated full-production yield (t/ha)


Nut characteristics

Size and shape

4.1g kernel.










Probably the most popular and promising of the USA imports.  Pistallate flower abscission (PFA) a major problem for this cultivar in California (due to over-pollination) but does not appear to be a problem in NZ.  In California, bunches of more than 3 nuts are very rare due to PFA but large bunches are common in NZ.

Some of the Serr in the Lincoln University trial block are almost certainly not Serr so treat the data very cautiously.  Serr will be trialled in the new 2005 NZWIG cultivar trial.

History: Bred in California (Payne was one parent) and was very popular in the 70s and 80s.  However at the time, PFA was an unresolved and poorly understood problem which made it go out of fashion by about 1990.  (In 1992, planted area was second to Hartley but new plantings were virtually none.)

Tree characteristics



Ultimate size




Blight resistance


Bud burst


Peak pollination


Peak receptivity


Bearing habit


Yield of 8–11 year trees (kg/tree)


Estimated full-production yield (t/ha)


Nut characteristics

Size and shape

3.1g kernel.


Very mild


76.5 (Highest equal with Rex)


Kernel thin and dark.




A Californian cultivar. It’s suitable for dry climates such as Central Otago, where it is grown for its big nut halves and good flavour. Frost is a significant issue for Vina and the main problem for the Central Otago growers.

Unfortunately, Vina is very susceptible to blight in all but the driest climates.  A heavy cropper with lateral bearing.  Tree has very poor apical dominance.  Heavy cropping appears to restrict its vegetative growth and its form is often bush-like.

In Central Otago, growers are using Hartley as a pollinator for Vina.

History: Bred in California (Payne x Franquette) and was very popular in the 80s until superseded by Chandler and more recently Tulare.  Import from USA.

Tree characteristics



Ultimate size



Very poor apical dominance

Blight resistance

Very poor

Bud burst


Peak pollination


Peak receptivity


Bearing habit


Yield of 8–11 year trees (kg/tree)


Estimated full-production yield (t/ha)


Nut characteristics

Size and shape

4.1g kernel. Thin pointed nut.


Mild, sweet




Kernel dark




A Californian cultivar, not widely tested in NZ to date. Tehama produces clean in-shell nuts.

In Central Otago, growers are using Hartley as a pollinator for Tehama.

History: Bred in California (Payne x Waterloo) and was mildly popular in the 80s.  Tehama was considered a bit of a ‘jack of all trades but master of none’ and lack of outstanding features meant that few were planted from about 1990 onwards.  Import from USA.

Tree characteristics



Ultimate size



Good apical dominance

Blight resistance


Bud burst


Peak pollination


Peak receptivity


Bearing habit


Yield of 8–11 year trees (kg/tree)


Estimated full-production yield (t/ha)


Nut characteristics

Size and shape

4.6g kernel.






Golden blonde




A Californian cultivar. It’s not widely tested in NZ to date but some very early orchards near Ashburton are growing Hartley and the trees are doing well.

The nuts are best for in-shell sales. The crackout rate is low, especially if they haven’t had enough water.

In Central Otago, growers are using Franquette as a pollinator for Hartley.

History: ‘Discovered’ in California in 1915 in the Napa Valley.  Most widely planted cultivar by far in California and still quite popular for new plantings amongst ‘traditionalists’.  Popularity is due to light kernel and resistance to blight and codling moth.

Tree characteristics



Ultimate size




Blight resistance


Bud burst


Peak pollination


Peak receptivity


Bearing habit


Yield of 8–11 year trees (kg/tree)


Estimated full-production yield (t/ha)


Nut characteristics

Size and shape

3.7g kernel.






Light blonde




An old French variety also popular in California.  Often used as a pollinator because it sheds its male pollen after most other cultivars so can assist in pollinating late flowering varieties Franquette is also very late leafing and has late pistillate receptivity.

History: ‘Discovered’ in France so long ago that no one can remember it not existing.

Tree characteristics



Ultimate size




Blight resistance


Bud burst


Peak pollination


Peak receptivity


Bearing habit


Yield of 8–11 year trees (kg/tree)


Estimated full-production yield (t/ha)


Nut characteristics

Size and shape

4.2g kernel.


Mild sweet




Clean, attractive shell. Kernel pale, plump, shiny.



CHAPTER 4 – Walnut tree density & layout

4.1 Maximising production

A walnut orchard reaches its optimum bearing potential when the space over the soil is completely occupied by nut-bearing tree canopies – as long as there is sufficient light penetrating to maintain lower fruiting wood. But, until the canopy closes, the production will be directly related to the number of trees you have planted.[1]

If your trees are initially planted at a high density for early production, it’s likely that you will need to remove some of them at 10 to 15 years to prevent crowding and shading reducing your production.  In planning your initial planting densities, it is essential to decide whether the extra production from the temporary trees will offset the costs of buying, maintaining and then removing them.

4.2 Recommended spacings

An accepted tree spacing for a mature orchard is approximately 10 m × 10 m but it will vary with the cultivar – Rex is a smaller tree than Meyric, for example.  Also, you should expect slower growth on a hard site with limited shelter than on a fertile, wind-free site, and you should take into account the rate at which the canopy will close.  Many growers aim for 10 m × 10 m spacing, or similar, but will initially plant at a higher density, and remove the temporary trees as the canopy closes.

Members of NZ Walnut Industry Group have observed a second thinning cycle in very old orchards (50–70 years) in France.  The final spacing after the cycle was approximately 20 m × 20 m.

[1] Walnut production manual, D.E.Ramos (technical editor), Division of Agriculture and Natural Resources, University of California, 1998.

4.3 Walnut tree layout

In preparing your planting layout you will need to consider both the density that suits you best and the probability that you will eventually want your harvesting done by machine. Straight rows, the same variety in each row and sufficient headlands between your nearest crop trees and your permanent shelter trees will make machine harvesting more practical.

Let’s look at some options for laying out a walnut block of size 80 m × 80 m (illustrated in Figure 4.1).

We will allow for headlands 8.5 m wide all the way round, to give us space for turning at the ends of rows. Some growers prefer wider headlands and 10 m is a popular choice, but in our example the size of the planted area becomes
63 m × 63 m, calculated this way:

80 m (block width) – 8.5 m (left headland) – 8.5 m (right headland) = 63 m

Here are some options for planting.

Planting at final spacing

A simple option is to plant trees on a grid spacing of
10 m × 10 m or similar.  At this spacing, it will be a long time before the canopy closes, but you will not have to remove any temporary trees.  (We haven’t shown that option in Figure 4.1.)

For a large, fast-growing cultivar, aim for 7 trees (i.e., 6 spaces between trees) across the block:

63 m (block width) ÷ 6 spaces = 10.5 m between trees

7 × 7 = 49 trees can be fitted into the block

For a smaller, slower-growing cultivar, aim for 8 trees (i.e. 7 spaces between trees) across the block:

63 m (block width) ÷ 7 spaces = 9 m between trees

8 × 8 = 64 trees can be fitted into the block

Planting high density and removing temporary trees

If walnut trees are initially planted on a 7 m × 7 m grid (tree positions shown as both P and x in Figure 4.1), it is possible to fit 100 trees (10 × 10) into the planted area:

63 m (block width) ÷ 7  m (tree spacing) = 9 spaces, i.e., 10 trees

7 m × 7 m spacing is reasonable for an initial high density planting.  The orchard can be managed either along the rows or columns of the grid, e.g., mowed along the line labelled M1in Figure 4.1.  Most growers spray out the grass and weeds along full rows of trees, so that they only need to mow up and down the rows.  Some growers only spray out patches around each tree, and would then mow both up and down and side to side.

After 10 years, and probably longer depending on the conditions, the canopy will be closing and it will be time to remove the temporary trees.  In Figure 4.1, the permanent trees are denoted P, and the temporary trees are denoted x.  If the temporary trees are removed, the orchard is left with a grid of permanent trees on a 9.9 m × 9.9 m spacing.  This grid is on the diagonal, as shown in Figure 4.1.  The orchard can still be managed along the original rows and columns (e.g. line the line labelled M1), or it can be managed on the diagonal.  You might choose to spray out the rows on the diagonal, and carry out orchard operations like mowing along the line labelled M2.

The final (diagonal grid) spacing can be calculated from the original grid spacing (7 m) as follows:

In this scheme there are 50 permanent trees in the block, and 50 temporary trees have been removed.

Growers often talk about their initial tree spacing being ‘10 by 10 with one in the middle’.  Figure 4.1 is one way of achieving that layout – the ‘10 by 10’ (approximately) grid is viewed on the diagonal, and there is a temporary tree in the middle of each grid square.  Another way of obtaining the same layout is to plant the permanent trees on a 9 m × 9 m or 10.5 m × 10.5 m grid, and then literally put the temporary trees ‘in the middle’ of each grid square.  This is illustrated in Figure 4.2.  It does not make a great deal of difference – the layout is equivalent to that of Figure 4.1 when viewed on the diagonal.

Of course, your tree layout does not have to be on a square grid – it can just as well be rectangular.

Figure 4.1: An example of high density planting initially, followed by removal of temporary trees at canopy closure. Permanent trees are denoted P, and temporary trees x.

If you are planning to remove temporary trees and/or change your management direction, take that into account when you design your irrigation system and install your irrigation sprinklers.  You won’t want to install sprinklers at locations that get in the way for later management operations.

Alternative layout for walnut orchard planting
Figure 4.2: A second layout for achieving ‘10 by 10’ (or other spacing) with one in the middle’. Permanent trees are denoted P, and temporary trees x.


Another layout option used in the USA and France is the hedgerow.  Trees are planted about four metres apart in the row, with about seven metres between rows.  This layout makes mechanical pruning possible – essentially side pruning with a hedge trimmer.  This is usually done on a four year cycle, with every fourth side (i.e., one side of every second row) being pruned each year.  Californian growers use the layout to manage the excessive vegetative vigour of walnut trees in their climate.  In New Zealand, the vegetative growth rate of walnut trees is very much slower, but hedgerowing could, in principle, be used for a different purpose.  The very high density planting achieves very early canopy cover, and therefore early bearing.  For mechanical pruning you would need to use one of the lateral bearing cultivars because they yield quickly from branches that have been headed back.  The main difficulty in New Zealand is that grafted trees are produced in relatively low numbers, so filling blocks at hedgerow spacing would not be easy or cheap.

CHAPTER 5 – Planting your trees

5.1 Equipment and preparation

Before planting your walnut trees, you will need to:

  • Mark out your tree rows and the positions for each tree.
  • Install your irrigation sprinklers and test them.
  • Suppress weeds around the planting sites.

5.2 Marking of the block with stakes or pegs

Working from your design, measure out the row and tree spacings in each block.  The spacings can be measured and staked along the shelterbelts, using a long tape measure or rope with distances marked on it.  Work along rows between stakes, marking each tree position.  Alternatively, you could ask someone else to help you.  One person stands at the stake by the shelterbelt, and judges by eye whether a tree position stake held by a second person is in line with the stake on the far side of the block.  Do it for each tree position, and mark the location with a stake or spray paint on the grass.  Measuring out is a slow job.

5.3 Ripping/sraying

It is usually easiest to spray out your walnut rows before you install your irrigation lateral pipes and sprinklers.  Suppressing the grass and weeds also makes digging holes and planting easier.

The usual way is to spray the full rows or, in an organic system, you would mow the weeds.

Some growers use a deep ripper along rows prior to planting, which can help break up soil that has been compacted during previous cropping.  Ripping also makes planting much easier if you are digging the holes by hand. It is probably unnecessary if you are digging deep wide holes with a auger or backhoe.

5.4 Timing

Walnut trees are notoriously difficult to propagate using the grafting method.  Consequently, each tree should be very carefully looked after before and after planting.  Propagators’ current practice is to plant nuts in August and grow them as rootstocks for two to three years.  They lift them in winter and graft scion wood from a selected cultivar onto the rootstocks inside a grafting house.  The graft must be kept warm for three to four weeks.  The trees are usually made available in spring or during the following winter when the tree is dormant.  Trees that have been kept in the nursery until the following winter will usually be more robust.

The root mass is typically quite large, and may have been pruned back.  Grafting tape may be still be wrapped around the graft.  Above the graft there will typically be two to three buds or shoots.  Take care when you are transporting and planting your new trees. Only handle them below the graft.

Growers generally plant after the risk of late frosts has passed – during mid-November in Canterbury.  New walnut leaf growth is very tender and susceptible to frosts, especially while the tree is young and close to ground level.  Immediately after grafting, the trees have very few bud sites on the scion wood, so having the shoots from these bud sites killed by frost is serious.  Fortunately, there are often two or more blind buds at each bud site and the scion wood can often sprout back from the second or even third bud at a site.  If the frost kills all the available buds, the scion wood will die, and the tree can only grow from the rootstock (below the graft) so you may as well replace it.

You may be able to buy second-year grafted walnut trees in winter.  Planting while the trees are dormant is a good option, and the trees are more resilient to frost during the following spring, because they have more bud sites.

In Central Otago bare-rooted trees are planted in September and trees from pots in mid-summer after the last frost.  Central Otago growers raise trees in areas protected from frost and don’t plant them until they are between two and two and a half metres high.  (If you were to buy walnut trees in Europe the propagator would only offer you trees of two and half metres because it would be a waste of money to plant them at the stage many are planted in New Zealand.)

5.5 Planting

Deep ripping the rows prior to digging will make the your planting much easier.  For larger numbers of trees, an auger or backhoe is the best option.  If you choose to use a tree auger, take the time to break up any glazed soil in the hole because it will restrict the growth of the new roots.

You will need an auger with a diameter of 400 – 500 mm and the holes should be made 400 – 500 mm deep.  A tree auger makes a good job of crumbling the soil into a fine tilth, which is easy to plant into, and makes it easier for the walnut roots to develop.  It is best to dig the holes on the day of planting so that the soil will not dry out.  If you must dig the holes earlier, re-wet the soil using the sprinklers of your irrigation system.

Check your plan to ensure that you have the correct cultivar in your hand, then place the root system at the bottom of the hole and gently push the loose soil around it, ensuring there are no air gaps.  When the soil is just above the root system, jiggle the stem up and down bringing the top of the root system up to just about ground level.  Gently place the rest of the soil around the tree and press down firmly with your hands.  Don’t stamp the soil down because it will compact the soil too much.  Remember, walnut trees like well-drained aerated soils.

Take care with your choice of fertiliser.  Some fertilisers will burn the new roots.  It’s usually best to spread the fertiliser on the surface after planting, if your soil analysis shows it’s necessary.  Well-rotted compost won’t damage the roots but make sure that it’s well mixed with the soil to ensure that you don’t have clumps that dry out quickly.

5.6 Protecting young trees

We recommend shelter guards for protection from:

  • browsing hares or rabbits
  • wind
  • herbicide sprays

Put the shelter guards around the trees on the day of planting. You will need to stake them firmly, but the soft soil will make it easy.

Finally, before looking back at all your hard work – years of planning and preparation – check that your nursery has given the walnuts their first application of copper spray.  If not, do it yourself to make sure that walnut blight does not get a hold in your new orchard

CHAPTER 6 – Designing and installing your irrigation

A useful reference on this topic is the New Zealand Irrigation Manual (described on  It’s also available at some libraries.

6.1 Water requirements

Irrigation will replace the water lost to evaporation and transpiration, commonly known as evapotranspiration (E.T).  In general, you will be irrigating both walnuts and shelterbelts through the summer – say, November to March, depending on your local climate and soils – but not in the winter.

You should design the irrigation system to provide for the driest months – January and February.  For that period, most designers use an E.T. figure of five millimetres per day.  An E.T. of 5 mm means that a volume of water equivalent to a layer 5 mm thick, covering your entire property, is being removed from the soil by evaporation and transpiration.  Likewise, if you receive 5 mm of rain, it is also equivalent to the same volume of water, so it will replace a single summer day’s E.T.  Many local newspapers provide the E.T. figures daily.  Try the weather page first.  The figures are based on a grass sward rather than on tree canopies, but can be taken as a useful indicator.  An E.T. of 4 mm per day is probably closer to the average, but it can be as high as 9 mm on a day of hot, dry winds.

In the early years you will probably only be irrigating patches around your walnut trees and along your shelterbelts unless you are aiming to grow grass between the rows.  However, you should design your irrigation system for the capacity your trees will need when they are fully-grown.  Once the orchard has a closed canopy, you will need to irrigate virtually the entire land area.  You can use the following simple formula to estimate the water you’ll need to irrigate the full orchard area:


Water requirement per day (m3/day)
= area to be irrigated (m²) x E.T. (m/day)


Area to be irrigated               =          15 hectares (150,000 m²)

E.T                                      =          5 mm/day (0.005 m/day)

So, water volume required     =          150,000 x 0.005

=          750 m3/day

To establish the required pump flow rate, divide the volume of water by the hours per day that the system could run.  It is wise not to count on running your pump for 24 hours per day. Allow for some downtime, for example a breakdown or power cut.

Say the system can be run for 18 hours per day.  Your pump must be able to deliver a flow rate of:

750 m3/day divided by 18 (hrs/day)    =          41.7 m3/hr

There are 1,000 litres in a cubic metre of water and 60 minutes in an hour, so we can convert the flow rate litres per minute:

41.7 m3/hr x 1,000, divided by 60       =          694 litres/minute (11.5 litres /second)

It is useful to have the required flow rate in both m3/hr and litres/minute as you speak to your regional council, bore drillers, irrigation engineers, neighbours etc.

You will need to determine whether any existing irrigation equipment on the block of land can supply the flow rate and determine whether local water supplies and regulations will allow you to abstract that flow rate.  If there is any doubt about obtaining the required irrigation water, you may want to engage an irrigation engineer to carry out a feasibility study and work out the cost before you purchase the property.

6.2 Designing and planning process

The main cost of system design and installation of a water abstraction mechanism (such as a bore and pump) will come early in your development but you may be able to spread some of the other items over the following years. If you are familiar with engineering calculations, you can probably design a large part of your irrigation system yourself, making use of resources like the New Zealand Irrigation Manual, but many growers engage an irrigation engineer to design their system. An efficient design will:
  • Optimise growth and maximise yields
  • Give you reliable operation
  • Provide an even application of water, at the required rate
  • Minimise the cost of pipework and other components
Be prepared to invest time and money to ensure the results are not left to chance or good luck.  You will need to provide the designer with the:
  • Location of your property
  • Soil type
  • Total area
  • Irrigable area
  • Crop types
  • Crop spacings
  • Crop numbers
  • Shelter layout
  • Available water source
  • A sketch or plan of proposed development.
    • Irrigation system components
The irrigation systems consist of four components:  pump, mainline, submains/laterals and emitter.  You will need to make some choices. Emitters: Drippers – usually used for shelterbelts and sometimes for young walnut trees. Sprinklers (large throw sprinklers, minisprinklers, microsprinklers) – usually used for walnut trees, at least after the first few years. Submains/laterals: Low density polyethylene pipe, or thin-walled lateral tubing Mainline: PVC pipe (rigid walled), or Polyethylene pipe (low or medium density) Pump: Surface – can draw water up from 6 m below surface or less. Submersible – can push water up from more than 6m below the surface.

6.4 Emitter selection

One of the first steps in designing the system is to decide on how you are going to distribute the water to the walnuts and shelterbelts.

Shelter belts

Shelter belts are best serviced with pressure-compensated drippers.  They are usually moulded plastic units, about the same diameter as a $2 coin.  They have a barbed entry suitable for punching into thin walled lateral tube.  Their intricately designed pathway and a rubber diaphragm ensure a uniform output.  If the water pressure throughout your system is relatively uniform, you may not need to go to the expense of the pressure-compensating dripper.  You will probably need them if your land is sloped.  It is worth trying to buy drippers that are ‘unblockable’, because some nozzles can be blocked by rust, sand, insects and seed heads, particularly during the winter.

Generally shelter trees are placed up to 1.2 m apart and one 4 litre/hr or 8 litre/hr dripper per tree is sufficient.  For widely spaced trees, use at least two drippers per tree.  The water emitted from the dripper will spread out in the soil to some extent, depending in the soil type.  The irrigated volume of soil will be reasonably large for a loam-type soil with good organic content, but will be smaller for a sandy or stony soil where the water drains through more quickly.

Walnut Trees

The most common emitter for irrigating walnuts in New Zealand is the microsprinkler which is very cost effective and efficient.  The ideal scenario is a modular type of microsprinkler that has a range of interchangeable nozzles and swivels that allow you to increase the irrigated area as the trees develop.

Most growers have a single microsprinkler per tree and place it about 0.5 m from the trunk, within the sprayed-out area of the drip zone.  In the first four years or so, they irrigate an area two to four metres in diameter, then change nozzles to irrigate a larger area.  Alternatively, you can add an extra microsprinkler per tree.  At that stage, most growers opt for sprinklers that will serve the orchard right through until tree maturity, so the new nozzles irrigate most of the orchard floor.  At all stages it is ideal for the irrigated area to be larger than the root diameter, to encourage the roots to develop.  However, irrigating a much larger area than necessary leads to wasted water and extra mowing.

Microsprinklers have a flow rate between 30 and 300 litres/hr, and the selection of their flow rate is an important part of designing your irrigation system.

Some growers make use of deflectors on their microsprinklers, so that water does not fall directly on the tree trunk or pond at its base.  It is unclear at this point whether deflectors are necessary.

If you have widely-spaced trees (e.g., 10 m ´ 10 m or more) a good option is the minisprinkler that provides an irrigated diameter of 14-16 m.  The sprinkler is still placed near the trunk to keep it out of the path of mechanical mowers, sprayers etc.  You’ll need a maximum of one minisprinkler per tree and you may be able to irrigate two to four trees with a single sprinkler.  Minisprinklers are more robust, have a greater nozzle size hence less filtration involved, and perform better in wind.  Whilst the initial cost is higher, the minisprinkler is well worth considering.

6.5 Mainlines, submains and laterals

In most irrigation designs, a large diameter mainline (e.g.,100 mm) runs from the pump through the middle of the property.  Submains then take water from the mainline out to walnut blocks.  Typically, the submains consist of 50 mm hard-walled PVC pipe, and they are usually laid along the shelterbelts on block boundaries (and left on the surface), or run along the block headlands (and buried).  Finally, lateral tubing of 13 or 16 mm diameter is ‘T-ed’ off from the submains.  One line of lateral tubing is run along each row of walnuts, and it should be mole-ploughed to a depth of approximately 100 – 200 mm.  The 4 or 5 mm tubes that feed the microsprinklers are plugged into these laterals.

The diameters and lengths of all the pipes are calculated during the design process.  If you are familiar with engineering calculations, you may be involved in the design process yourself.  A good rule of thumb is that the velocity of water flow through the pipes should be about 1 metre per second.  If you find that the velocity is lower than 0.5 m/s in any section, your pipe may be unnecessarily large at that point and thus unnecessarily expensive.  If the velocity is greater than 1.5 m/s in any section that is longer than 10 metres, there will be too much pressure drop in that section and you should consider increasing the size of the pipe.

You will also need to decide what area of the orchard you will irrigate at a time.  Except for very small orchards, it is usually not possible to do it all at once, so designers split the property into irrigation blocks.  Each irrigation block may consist of a single walnut block, or several blocks grouped together.  If you have several different soil types on your property, it is sometimes possible to include the areas of primarily light soil in different irrigation blocks from those on primarily heavy soil, so that you can irrigate them for different durations or at different frequencies.

Place a control valve on the submain leading to each irrigation block, so that you can control it independently.  Irrigation to the shelter should also be controlled independently.  The best choice of valve is electric solenoid control via a 24 volts AC wire system feeding back to a simple irrigation controller in your garage or pump shed. The control units make it easy to operate the system automatically.

6.6 Pumps and bores

You will need some essential information before you can select your pump:  the flow rate and the pressure lift.  You can calculate the flow rate using the example in the ‘Water Requirements’ section.  The pressure lift takes into account the distance that your pump must lift the water and the pressure you need at the wellhead to run your emitters.  (You may need some help from an irrigation engineer to calculate this.)

Pumps are generally driven by electric motors.  With today’s technology, they are quiet, clean and efficient, but a submersible pump is expensive and you’ll probably need professional help to select the right model and install it down your well.  You will need a pressure in the mainline of about 350 kPa.

The filter is an essential component of the system.  Choose one that matches the flow rate of the system and the type of debris you need to filter.

If you are having a bore drilled it is very important not to compromise on the components to cut costs.  You want your bore and pump to have many years of trouble-free life, running many hours per day, and it is worth getting everything right the first time.  It is a very expensive exercise to have your pump pulled to the surface for maintenance.  The bore must have a stainless steel wedgewire screen at the bottom, and should be developed with the well driller’s submersible pump (not yours!).  Development of a bore is the pumping out of sand and debris from the immediate vicinity of the screen where the water is drawn in.  The idea is to leave only clean gravel of diameter larger than the screen mesh so that you will not be pulling in sand with the water.  Developing will preserve the life of your submersible pump and prevent stressing or overloading the filtration system.

Another cost which is often overlooked in the budget is the reticulation of power (often three-phase) to the pump site and the electrical starter and safety protection equipment for the pump.

It is cheaper to reticulate water than power, so if water is accessible anywhere on the property choose a site that is as close as practicable to the three phase power.

6.7 Installation timeline

The pump and bore should be installed first, followed by the mainline and some or all of the submains, depending on your planting schedule.  It is usual to plant shelterbelts in the winter months.  You can then lay lateral tubing along the trees and plug in a dripper for each tree, ready for the dry summer months.

It is usual, however, to install lateral tubing for walnut irrigation before the walnuts are planted.  The tubing is mole-ploughed along the walnut rows to a depth of 100 to 200 mm, and microsprinkler leads are plugged into it while the soil in the rip line is still soft enough to push aside by hand. It is important to set up the irrigation in time for the walnut planting because you may be planting into dry soil and your trees will need irrigation right away.

Table 6.1:  Installation timeline

Year 1Year 2 or 3
BoreLateral tubing for walnuts
Power to siteMicrosprinklers in first blocks of development
Submersible pumpSupply tank and domestic pressure
ControllerDomestic pump (if required)
Filter assembly 
Mainline, submains 
Shelter irrigation 


6.8 Concluding comments

As you design and install your irrigation system you will find a wealth of expertise and experience available from walnut growers, regional council, well drillers, electricity suppliers, irrigation engineers, horticultural advisers and neighbours.

The average cost of an irrigation system, not counting the well or power reticulation, is between $3,000 and $4,000 per hectare, so it’s worth spending time in the planning stages and learning from others.

Even if you have installed the most technically advanced irrigation technology, you remain the most important part of your scheduling system.  It’s vital to monitor your moisture and schedule your irrigation carefully.  We must all contribute to managing our precious, finite water resources and optimising its returns.  Careful management will enable our children and grandchildren to have the same quality of water that we so often take for granted.


CHAPTER 1 -The early years

In the early years you should aim to create maximum vegetative growth and a large fruit-bearing structure of suitable form.

You need to manage anything that impedes vegetative growth, including suckers, reproductive effort, competition from weeds and grass, frost damage, wind stresses, inappropriate water levels and pests such as hares and rabbits.

This chapter will help you with tasks that are specific to managing walnut trees in their first two years.  Weed control and irrigation management are important in the first two years, as well as being on-going management tasks.  Young trees are less able to compete with grass and weeds for water and nutrients.

1.1 Removing suckers from below the graft

Grafted trees tend to grow suckers (shoots grown from buds) on the rootstock below the graft.  Remove them as soon as possible so that the plant can direct its energy above the graft where you want it.  During late spring, when the vegetative growth rate is high, remove the suckers on about a three-week cycle.

Rub emerging buds (below the graft) off with your thumb. Break small suckers off by hand and cut larger ones with secateurs.  Take care to minimise damage to the bark and cambium.

Removing suckers is most important in the early years because they compete strongly with the scion.

1.2 Removing nutlets and catkins

A tree’s energy is channelled into a combination of vegetative growth and reproductive effort.  During the early years you want the tree to focus on establishing a fruiting canopy, so remove nutlets and catkins.  Catkins are visible and easy to detect and remove in the winter and you can remove them easily by hand.  Pistillate flowers, which become nutlets, start to appear in the first half of October and can be removed by hand at that stage.  However with all the vegetative development during spring it is not always easy to detect the flowers, so it pays to revisit in November when the nutlets are formed.

1.3 Protection from pests

Rabbits and hares can cause significant damage to small walnut trees.  Hares tend to bite them off completely soon after they are planted, perhaps because they like to maintain clear visibility in their territory and want our ‘obstacles’ out of the way.  However, they do not seem to damage walnut trees once they have accepted them as part of their territory.  The damage potential from hares is usually less if the area is already covered in long grass, because the hares’ view is already obstructed.  Rabbits need to regularly grind their teeth and are quite happy to make use of a young walnut tree for the purpose.  They may occasionally ring-bark a tree, but often the damage is just enough to weaken the tree’s structure so that it is broken in the next strong wind.

Birds, in particular magpies, can cause some damage to small trees.  Magpies have a habit of perching on fragile immature branches of walnuts trees and breaking them off with their weight.

Use of shelter boxes and spray guards

Shelter boxes will protect your trees from rabbits and hares, wind and drift of herbicide sprays. The 750 mm high boxes are probably the best.  The cheaper, shorter ones are effective in most cases but at least one grower has evidence of acrobatic animals reaching over them.

Square boxes of width 100 mm ´ 100 mm are popular and provide sufficient space for small trees.  The box material allows virtually all the important wavelengths of solar radiation to reach the trees inside.  Check the size of the tabs and height of the box and buy stakes to fit them.  The stakes are usually wooden and it is important to use a ground-durable timber or suitably treated timber – untreated pine lasts only 2 to 3 months under irrigation!  Shelter boxes cost about $2.50 each, plus 70c for the stake.  They have a lifespan of 5 years or more so it may be worth buying them secondhand.  If you are buying them new, you might be able to recover some of your cost by selling them when your trees have outgrown them.

Stakes can break in strong winds, particularly if there is a large ‘sail’ area of leaves above the box.  Therefore it is worth inspecting your trees during or immediately after a strong wind.

Shelter boxes are widely used in Canterbury, but growers in Central Otago find that in the hot summers the trees can overheat so it’s best to add some holes in the bottom for ventilation. Some growers are concerned that the boxes give too much support and don’t encourage the trees to strengthen.

Green plastic sleeves cost about 30c each and provide protection for herbicide spray drift and some protection from pests but are not as effective as shelter boxes.  Pest-deterring pastes are used for low value trees such as pines.  They may provide some protection but are not a total pest management system for orchards with a significant threat from pests.

The usual approach is to protect young walnut trees with shelter boxes for the first year or two until trees are less vulnerable to pests.  You can then replace them with spray guards to protect them from drift of herbicide sprays for the next couple of years.  Large trees often start to rub on their boxes, inflicting damage to the bark.  Such trees are usually large enough to withstand wind and pests, and it’s best to remove their boxes.  Removing the box causes a dramatic change of environment (and protection from wind) for a tree.  The ‘optimum’ time to remove boxes from smaller trees is hard to determine.  They should be removed as soon as the tree can withstand wind and pests but may be left on for a second growing season.  It is easier to remove boxes and stakes in the winter when the ground is soft and the trees have no leaves. You may also minimise the shock.

1.4 Protection from frost

Young trees are susceptible to spring frosts after bud-burst in many parts of New Zealand. The most vulnerable time is immediately after spring planting, but young trees can also be frosted in subsequent springs, with the risk reducing as the tree grows to more than three metres.

Frosts are seldom severe enough to cause damage during winter dormancy.  In many parts of New Zealand ‘inversion layer’ frosts are common.  The frost effects are greater closer to the ground, and often there is little or no effect two metres above the ground.

Frosts can kill all growth from a bud that has burst.  The trees will then usually re-sprout within a couple of weeks from buds that had not burst at the time of the frost.  Often a bud site contains two or three buds.  In normal circumstances, the top bud bursts first, but if it is frosted the next bud will burst so that even when all the foliage appears to have been killed by frost the tree will usually survive.  After a frost, leave the tree to regrow, and do not prune until the danger of a damaging frost is over.  A copper spray may help prevent blight entering through the damaged tissue.

Newly planted trees are most susceptible to frost damage because they have the fewest buds.  If all viable buds on the scion wood have burst and are killed by frost, the scion wood will die.  It is unlikely that frost will kill a second year (or older) tree because there are usually sufficient ‘reserve’ buds.

Copper sprays (Kocide in particular) may provide a small level of frost protection and may prevent damage from a light frost (-1 to -2ºC).  The protection is apparently due to the reduction of bacteria on the foliage.  The bacteria would otherwise provide nucleation sites for ice crystals.  A simple way to protect young trees in shelter boxes from frost is to cover the top of the box with a piece of frost cloth (doubled over), using hat elastic to secure it.  You could even use screwed up newspaper in the top of the box.

Covering insulates the tree’s environment and maintains it two to three degrees Celsius warmer than without the cover but it will not provide protection from a heavy frost.  You can leave frost cloth on small trees in boxes until they reach the top of the box.  Alternatively you can cover small trees with double-walled paper sacks or potato sacks etc.  The sacks appear to give better frost protection, but don’t leave them on for more than a few days, or the tree will suffer from lack of light.  The paper sacks are also susceptible to wind or rain.  For small plantings, 44 gallon drums (with lids) placed around the young trees provide good frost protection.

CHAPTER 2 – Managing the orchard floor

2.1 Introduction

In this chapter, the ‘orchard floor’ includes both the upper layer of the soil and any vegetation cover on it.

For management purposes, we can think of the orchard floor as two areas:

  1. The drip zone: the area described by the mid-day shadow of the tree, or the area in which water would drip from a wet tree.  It is the main area from which the tree roots are feeding, and where the nuts will fall.  Note that for young trees in training with the modified central leader system, the roots may extend further than the above-ground foliage.
  2. The service area: the rest of the orchard.

The proportions of each area will change as the orchard grows.  When the trees reach full canopy cover, the service area will be restricted to the orchard borders and access alleys.

2.2 Managing the drip zone

The drip zone is the top priority in orchard floor management.  There is no single ‘correct’ way to manage the drip zone but we can suggest some key issues to help you decide what you want to achieve and describe some techniques of management.

What do you want to achieve?

Unchecked weeds in the drip zone will constrain the growth of your trees as they compete for water and nutrients, and may obstruct your irrigation sprinklers.  However, a ‘bare earth’ policy is expensive in consumables and labour and may not be totally necessary at all times.  Let’s look at the issues you need to consider when deciding how to manage your drip zone.

Nut Harvesting

It is more efficient to harvest nuts from a flat surface with minimal vegetation.  Long vegetation tends to be damper and increase the rate of degradation of fallen nuts.

Avoiding interference with irrigation systems

Micro-sprinkler irrigation systems are only effective if the immediate area around the sprinkler is free of vegetation at the height of the sprinkler.

Competition (for water and nutrients) with the walnut trees

Vegetation will compete with the walnut trees for water and nutrients in the topsoil.  If water and nutrients are available, the walnut tree will gain about 95% of its needs from ‘feeder’ roots in the top 400mm.  If vegetation out-competes and depletes the water and nutrients in this top zone, the tree will have to feed deeper and overall its uptake will be reduced.

Weeds can be gross feeders and it is unlikely that a walnut tree could compete against dense weed cover without losing some of its growth.  Conversely, a low-density weed population is unlikely to have a significant impact on walnut tree performance.

There are ways in which vegetation and trees can co-operate rather than compete; the classic example is clover and the availability of nitrogen.  Clover is able to ‘fix’ nitrogen (taking it in from the air through its leaves and depositing it in an available form in the soil) to the benefit of trees.  You would need to allow for the clover competing for moisture.

Soil Health

Although there are many disadvantages in having grass/weeds in the drip zone, there are also some benefits to soil health in maintaining at least some vegetation cover.  This cover could be in the winter only, or at a controlled level during the growing season.  In particular, there are significant benefits to the soil from the added organic matter.  Decaying vegetation (both foliage and root biomass) will improve the soil’s ability to hold moisture and nutrients.  It will also attract worms and other soil fauna that improve the mobility of the water and nutrients.  However, even in bare soil, some worms will probably be active as long as you keep it moist.  Deep rooting vegetation provides ‘nutrient cycling’ – translocation of nutrients from the deep parts of the root zone to the top part of the root zone, adding to the nutrient cycling already performed by the deep-rooting walnut trees.

A layer of vegetation also protects the topsoil from wind erosion and surface structural damage that can occur when rain or irrigation water falls directly on the surface.

Simplifying service area management

Some growers manage their orchards in full rows (e.g., spraying out the grass along a full row of trees), and others manage a drip zone area under each individual tree (e.g., spraying out patches under each tree).  It is much easier to mow your service area when there are only strips. You will need to mow in two directions if you choose to spray only the patches under each tree.  However, the latter option requires less herbicide usage.

How to manage vegetation in the drip zone

Herbicide spraying is the usual tool for managing weed and grass growth in the drip zone.  You may prefer to use mulch, particularly if you are growing organically, and mulch has the added advantage of conserving the moisture in the soil.  You may read of cultivation as a weed control method in old American books.  We don’t recommend it for the drip zone.  (See Section 2, Orchard Floor Planting, for the case against surface cultivation.)

Spray basics:  types, application rates

There are essentially two types of herbicide spray:  foliar sprays and pre-emergent sprays.  Foliar sprays kill following direct contact with foliage.  They are either systemic sprays (i.e., the chemical moves through the whole plant system from any spray contact point) such as glyphosate and Versatil, or non-systemic sprays, such as Buster, which simply burn the leaves.  Pre-emergent sprays bind to the soil and kill the vegetation as it germinates.  Because they are residual in the soil for a period, they will continue to control vegetation growth after you have sprayed.  Simazine is strongly residual (and thus effective for a longer period, depending on application rate), and Foresight less so (as it only forms a layer on top of the soil).

Of the systemic foliar sprays, some are residual in the dead vegetation and soil for shorter periods than others.  Glyphosate quickly becomes inactive in the soil, but it is important not to allow it to come in contact the bark of young trees.  With any of the foliar sprays, you will need to re-spray each time a new flush of weeds and grass appears (usually two to three times per season) whereas the pre-emergent sprays can be effective for a whole season.  Moreover, some spray drift from a pre-emergent herbicide will not harm the walnut trees.

It is well worth adding a penetrant to foliar sprays.  You can use Pulse, at the top end, or dishwashing liquid, which is somewhat less effective but much cheaper.  Without a penetrant, it can be hard to get the spray to stick to the vertical grasses and the hairy and waxy weed leaves.

Spray concentration is the ratio at which the spray is mixed with water and application rate is the volume per hectare (either volume of active spray or dissolved volume at a given concentration) applied.  Always read the label carefully.  If you use the chemicals according to the manufacturer’s or distributor’s specifications and conditions, they will do what is expected in virtually every situation.  The suppliers want to strike a balance.  They need to recommend a rate and practice that works all situations, so every customer is satisfied, but they also want the application to be cost effective against competitors, and are generally interested in safe and sustainable use of their chemicals.  So the label will state a range of rates, and generally, if conditions are right, the lower rate will be enough.  You may even find that slightly less than the manufacturer’s lower rate will do the job.

Choosing a management option

You now need to decide on the management plan you will use to achieve your goals.  We suggest two possible options.  The first aims to keep the drip zone free of weed and grass growth for virtually the whole year.  Many growers who are harvesting walnuts from mature trees choose the weed-free approach.  The second option is a compromise between optimising walnut tree growth and needs of the soil ecosystem.  You would spray out most of the weed growth during the spring and summer, but allow vegetation to return while the trees are dormant.

Drip zone bare year-round

Kill any existing weeds during the winter (with a foliar spray in late autumn/early winter and another in late winter), and then, in the early spring, apply a pre-emergent spray for control over the summer.

It is probably a good thing to see a little weed break in the autumn as it suggests your residual spray is just reaching the end of its effective life in the soil and is not accumulating.  It is better than simply applying full rates each year and possibly wasting money and building up chemicals in the soil.

In addition, these perennial weeds and grasses in the autumn, while not going into full dormancy, are in fact drawing reserves down into the plant before the winter, so early winter (about May) is a good time to make use of this process with a foliar spray.  Take care to minimise spray drift onto walnut trunks and remaining leaves, particularly if using glyphosate, as any significant contact will show up in the spring growth.  The spray will be absorbed and translocated, and not broken down in the cool winter conditions.

After spraying in late May, you will probably notice a flush of grass and weed seedlings over June or July.  Those seedlings are very susceptible and can be removed by going through quickly in early August with a low rate of foliar spray.  It will clear your orchard of vegetation, ready for spraying a pre-emergent herbicide in the spring.

By late August/early September, the late winter spray will have created a clear soil surface. The ideal is to apply the pre-emergent spray just prior to the spring weed strike.  It is best to spray onto damp soil, so that it will spread.  However, try not to spray just before heavy rain or it will not have a chance to attach to the soil particles, and could wash down to the walnut roots, or worse, into the sub-soil.  If your soil is relatively light (i.e., largely silt and not too much clay or organic matter), and you are in a fairly low rainfall area, you can use the low end of the recommended rate. The spray should provide effective control throughout the summer growing season.

Drip zone weed-free only during growing season

If you choose this option, you won’t use pre-emergent sprays.  You can knock down grass and weeds most effectively by foliar sprays while they are growing actively.  It’s important to spray in early spring, as soon as grass growth begins in earnest in September.  How often you spray after that depends on your desire for tidiness, and clear ground under the trees.  You will probably need at least one other spray around November, and possibly one in February.  Once your harvest is over, and the trees become dormant, there is little need for controlling grass and weed growth – unless you plan to use a pre-emergent spray in spring.

Glyphosate at low rates will kill off grass and many other weeds, but only knock back clover. Then, in the absence of competition the clover recovers and flourishes.  Thus if you are aiming for nitrogen-fixing clover in the drip zone (also providing a protective cover on the soil, and flowers for bees and other beneficial insects), this is a good way to achieve it.  You can also knock the clover down just before harvest with Versatil.  You will then be harvesting from a flat, clean cover of dead leaves and stalks rather than from bare soil.  Make sure you achieve a good knock down, otherwise the clover will cause problems for picking up the nuts.

Spray equipment

Typically, growers use a knapsack sprayer when the trees are small, and later a boom sprayer on a tractor, quad bike or trailer.  Many growers with smaller orchards have a contractor do the spraying at this stage, so that they do not have to purchase all the equipment.  The decision will also depend on your time commitments, whether you own a tractor, and whether you are spraying full rows or just the drip zone under individual trees.

You use less spray with a knapsack, and can direct it where it is needed, but it does take more time.  One option is to mix a large tankful of spray and take it into the orchard for re-filling the knapsack, since the time consuming part of knapsack spraying is usually walking back to the shed and mixing the spray.

When you are spraying with a knapsack, it is possible to apply different rates to different plants.  For a dense and/or difficult weed or vertical grass, you can pause and apply more spray, until it is running down the leaves into the growing area.

Young trees should either be in shelter boxes or have a spray guard around their base to protect them from spray drift.

The drip zone in mature orchards

Once the walnut trees are large, the combination of shade and fallen leaves discourages the clover, as well as most other vegetation.

Consider using smaller amounts of a wider range of chemicals in your spraying.  Plant species vary in their susceptibility to particular chemical sprays.  Some plants will be totally controlled, even with low rates of spray, some will be fully resistant, and the rest will be somewhere in between.  So if you use only one spray type, the weeds resistant or semi-resistant to that spray will recover and then flourish because of reduced competition.  One option is to apply one spray broadly and then come behind and spot-spray the invaders where necessary with a different spray.  You could mix them together but you’ll need to be especially careful with some mixes.  Check the instructions or talk to your supplier about the safety of mixing your chemicals.

Managing the service area
Maintaining the strips between trees or tree rows as a mown grass/clover sward is the easy norm.  Cut grass and clover can be mulched back into the soil, so that nutrients remain on the property and you increase the organic content of your soil.  However, there are also a few other options.  Some growers use orchard grass – the low, creeping lawn grasses such as timothy and fescues – to minimise mowing.  They are expensive, and other grass species often creep in, but may be an option for smaller orchards.

There may also be other options for the service area.

Growing shelter for young walnut trees

We often have inadequate shelter when the trees most need it (when they are small) and too much when they are mature which leads to frost ponding and high humidity in the orchard.  One option is to use the service area to create some shelter at the low level while the trees are small.  The first seemingly obvious choice is sweet-corn, but the corn may not be big enough in spring, when you need it most, and there is no other annual crop tall enough.

The best option is to establish your shelter two to four years before you plant your walnuts but many growers plant smaller blocks and fast growing, possibly temporary, shelterbelts.

Productive crops

Because of the very long lead time before you begin to derive an income from your walnut orchard, a complementary shorter-term crop between the trees may seem like an attractive option.  However, take care in your planning, since it can be difficult to manage two separate crops in the same area.  Even so, there are some options worth considering.

One or two cuts of hay or silage could be harvested each year before you plant your walnuts – or even from between the walnut rows.  Your local farmers may help you select a suitable pasture mix for your growing conditions.

There’s a complication with hay or silage.  It is becoming increasingly difficult to find anyone to cut and bale hay or pick up silage on small blocks.  You should also consider the risk to your shelter and walnut trees from having large machinery working around them.  If you have your own small baler you could do the job yourself and sell the bales.

Vegetable or herb crops could certainly be grown between rows, as long as you can arrange suitable irrigation and they do not obstruct access to your walnut trees.  These options may be practical for a smaller orchard where more tasks are carried out manually or with smaller equipment, but may be more difficult in a large orchard where many operations are carried out by tractor.

At present in the New Zealand walnut industry, the majority of growers work off-farm while they are establishing their walnuts and very few have the time for complementary crops. One option is to lease the land to a grain or vegetable grower while you wait.


Grazing animals may damage young trees and even if your trees are already mature, the manure will make harvesting unpleasant, inefficient and unhygienic.

At harvest time

Mow the grass very short in preparation to harvest. You may need to cut it several times in the lead-up to harvest so that the grass is not clumpy or stalky.  In most regions it’s necessary to have the orchard floor tidy before the end of March.

CHAPTER 3 – Nutrients

Walnut trees perform best if the soil nutrient balance and pH are within certain ranges.  Table 3.1 describes key elements within the soil, and Table 3.2 provides ideal nutrient levels for walnut foliage.  The usual approach to nutrient management is to sample and test the soil before planting walnuts, broadcast fertiliser as required to address any major imbalances, and then to annually monitor the walnut trees using foliage testing.  Particularly once you are harvesting walnuts, some nutrients will be leaving the orchard in the nuts, so you will need to replace them with either ground based or foliar fertiliser.

You should use your soil or foliage testing and fertiliser regime to maintain your nutrient concentrations within the specified ranges.  It’s also important to monitor several important ratios of one element to another and maintain them between upper and lower limits. Use the test results to complement your visual monitoring of the health of the trees.

Maintain the organic content of your soil so that it will have good water- and nutrient-holding capacity.

Be careful with irrigation, particularly on heavy soils.  Try to keep a healthy, aerated soil under the irrigated area.  Worms can be both tools and indicators.  Their activities aerate the soil and improve drainage.  Worms prefer aerated, non-acidic soil, with good structure and plenty of organic matter, which is also a good soil for root growth and tree health.

3.1 Soil sampling and testing

Several commercial laboratories provide a nutrient analysis service.  Most will supply sample collecting instructions and bags for sample storage and dispatch.  You can collect the sample yourself and send it to the lab.  The laboratory will provide a detailed report of nutrient status.  You then need to interpret the results and determine the mix of fertilisers required to address any deficiencies.  Some laboratories will explain the significance of the results and supply a recipe for a suitable fertiliser mix, or a horticultural advisor can help you with this.

If there is any reason to suspect that the nutrient balance in different parts of the property will be different, because, say, the soil types vary or they have a different fertilisation history, you should sample and test them independently.

Sample with a ‘pogo-stick’ – a sharpened stainless steel cylinder which is trodden into the soil and provides cylindrical cores of approximately 100 ml volume.  Horticulture supply stores, Ravensdown, Agriculture New Zealand and some laboratories supply them, but you can make your own from tubing.  A sample consists of a set of cores; the cores should be collected from the whole area to give a representative sample.  Send the samples to the lab (in the sample bag provided by the lab) with minimal delay.  The results are usually available in a week or so.

3.2 Interpretation of soil test results

Your soil test results describe some of the basic characteristics of the soil (such as pH and Cation Exchange Capacity – the soil’s acidity and its ability to hold nutrients) and measure the nutrients available to plants (Olsen phosphate, and levels of potassium, calcium, magnesium and sodium).  You may also receive assessments of other characteristics such as organic matter.

Most laboratories compare your soil nutrient levels with ‘low’, ‘medium’ and ‘high’ ranges (or ‘below’, ‘biological optimum’, and ‘above’ ranges) so that you know whether you need to take action or not.  The DSIR soil bulletins define ‘high’ nutrient concentration to mean a level in excess of that required for maximum production – so if you add more of that nutrient in a fertiliser, you don’t expect a response in plant growth.  A ‘low’ nutrient concentration is less than that normally associated with maximum production, so you will probably get a plant growth response if you add the nutrient in fertiliser.  In the middle you have ‘medium’ or ‘optimum’ nutrient concentration, which is associated with maximum yield.

Don’t allow nutrient concentrations in your soil to go below the ‘low’ range into the ‘deficient’ range or above the ‘high’ range into toxic levels.  However, to get the full picture, you should interpret soil nutrient results alongside leaf test results and the general health and appearance of the tree.

Be careful if you are comparing soil nutrient levels from two different laboratories, as results can be given in several different units (e.g., MAF QT, ppm or me/100g).  It is possible to convert between the different units if you know the soil’s bulk density, or if the two results are for the same soil.  Generally, though, you don’t need to worry about converting – just compare your soil results with the ‘low’, ‘medium’ and ‘high’ ranges provided by your laboratory.  These results tell you the total amount of exchangeable calcium, magnesium, potassium and sodium.

Table 3.1:  Key soil characteristics

Note that we haven’t provided a ‘medium range’ for all characteristics because the measurement units vary between laboratories. Make use of the ranges provided by your laboratory

CharacteristicMedium rangeComments
pH (acidity/alkalinity)6.0 – 6.5The availability of nutrients can vary according to pH.
Cation Exchange Capacity (CEC) = Cation Storage Capacity (CSC)

11 – 25 me/100g

(me/100g is the same unit as me%)

A measure of the soil’s ability to store Ca, Mg, K and Na.
Phosphorus (P)20 – 25 mg/mlOlsen P test measures immediately available phosphorus.
Sulphur/Sulphate (S/SO4)

10 – 15 ppm

(ppm is the same unit as mg/g)

Immediately available sulphur reserves.
Organic Sulphur

15 – 20 ppm

(ppm is the same unit as mg/g)

Slowly available sulphur reserves.
Calcium (Ca) Available calcium – low levels may limit earthworm activity.
Potassium (K) Immediately available potassium.
Magnesium (Mg) Immediately available magnesium.
Sodium (Na) Available sodium.
K/Mg ratio0.3 – 1.0 
Total Nitrogen (TN) The portion of total nitrogen that is actually available to plants depends on temperature (so it is seasonal).  It also depends on the OC:TN ratio – ratios higher than 12:1 (e.g., 15:1) can limit N availability.
Organic carbon (OC)

> 4%

Organic matter (OM) = 1.7 ´OC, so OM > 7% is ideal.

Organic matter improves soil’s ability to retain nutrients and moisture.  But also keep OC:TN below 12:1 (e.g., 10 parts OC to 1 part TN is fine) so that N is available.

(Information extracted from DSIR soil bulletins, discussion with Hill Laboratories and soil analysis reports of Celentis Analytical).

Your results sheet may also include ‘base saturation’ values, which tell you about the relative availability of calcium, magnesium, potassium and sodium.  Unfortunately these results, also, are not always expressed in the same way by different laboratories.  Table 3.1 describes some of the results that you may receive from your laboratory, but does not provide ideal ranges for all cases, because they are often specific to your laboratory.

More detailed information on interpretation of your test results can be requested from your testing laboratory, or from a fertiliser supplier.

3.3 Leaf sampling and testing

Several commercial laboratories offer services supplying sampling bags and analysing leaf samples.  Take the leaf samples in the last week in January or early February from healthy, mature leaves from typical trees.  Select leaves from half way along the current season’s growth.  You’ll need to take leaves from around 25 trees for your sample.  If there is any reason to suspect that the nutrient balance in different parts of the property will be vary, sample and test each area independently.

Keep the samples in a dry place (not the fridge) and send them to the lab as soon as you can.  Expect the results in about a week.

Note that the concentrations of chemicals in the leaves are different from those in the soil and there are different recommended ranges for results for soil tests (Table 3.1) and leaf tests (Table 3.2).  Either your laboratory or a horticultural advisor can help you design a suitable fertiliser programme based on your nutrient results.


Table 3.2: Recommended nutrient levels for leaf testing

CharacteristicRecommended range



Nitrogen (N)2.5–3.5%The most essential element.  Deficiencies cause slower branch growth, yellowing of leaves.
Phosphorus (P)0.15–0.3%Required for energy conversion and root development.
Potassium (K)1.0–3.0%Boosts immune system.  Deficiency causes discoloured patches on leaves and increased fungal infection and insect attack.  Excessive potassium levels make magnesium unavailable to the plant.
Sulphur (S)0.15–0.40%Boosts immune system
Calcium (Ca)1.2–2.5%Boosts immune system and helps in small root development.
Magnesium (Mg)0.3–1.0%Critical for photosynthesis, hormone activity and many other plant functions. Excessive magnesium levels make potassium unavailable.
Sodium (Na)<0.1% 
Iron (Fe)20–100 ppmImportant for nutrient transport within the tree.
Manganese (Mn)30–300 ppmImportant in cell division and hormone activity.
Zinc (Zn)20–200 ppmRequired for good shell development in nut trees.
Copper (Cu)4–20 ppmRequired for cell structure.  Copper sprays for blight usually supply all that is needed.
Boron (B)35–300 ppmThe most important micro-nutrient for walnuts.  Required for all tree functions including producing nuts.

Leaf nutrient level information compiled from several laboratory sources, Ross Jamieson and the Californian Walnut Production Manual.

3.4 Fertiliser programme

Your fertiliser programme should address any major nutrient deficiencies or imbalances you find in the leaf or soil analysis and observations of the health of individual trees.  It is not worth investing heavily in fertiliser for minor issues.  Testing in subsequent years will indicate whether the fertiliser has corrected the deficiency or imbalance.  Graph the leaf analysis results each year to help you track the changes but keep in mind that different fertilisers have response times that could range from less than one year to five or more years.

You will probably need some help from a specialist to design an effective fertiliser programme, at least in the first couple of years.

The organic content of your soil determines how well it is able to retain moisture and nutrients and how available they are to the tree.  Earthworms and soil micro-flora prefer a soil with high organic content and in turn they improve the soil’s structure and drainage and make the nutrients more accessible to your walnuts.  The best way to improve the organic content of your soil is to grow plants in it.  A vigorous grass sward will add a significant amount of organic carbon to the soil from the roots and also from the leaf material if you don’t take it off as hay or silage.  The alternative of mulching will cost you time and money.

Another option is to provide additional nutrients through your irrigation system.  You can use the irrigation for organic growing too, provided you use the approved fertilisers.

CHAPTER 4 – Disease

4.1  Walnut blight

Walnut blight is the most significant disease affecting commercial walnut orchards in New Zealand. It can attack and destroy nuts, and, in the worst cases, can cause almost complete crop loss.  It can also cause significant reductions in vegetative growth. 

Despite extensive searches, particularly in the USA and France, no cultivars have been found which are wholly resistant to blight.  Of the New Zealand selected cultivars, Rex appears to be the least susceptible to blight (at least in the drier, eastern climate), while Dublin’s Glory is one of the most susceptible. 

A warm, wet or humid spring provides ideal conditions for blight growth. In areas where those conditions are common it is virtually impossible to manage blight effectively.

Visual symptoms

Walnut blight is caused by the bacterium Xanthomonas campestris pv. juglandis, and it occurs only in the Juglans (walnut) genus.  It can attack current season’s soft green growth – the shoots, leaves, catkins, pistillate flowers and nuts.  Growth from previous seasons is not susceptible.

Watch for black lesions on nuts or growth shoots. (You’ll find examples on ‘The Problem Page’ of the coloured picture section.)  It is easiest to find on nuts where any black spots greater than 3mm diameter are very likely to be a symptom of blight.  Large black spots on shoots may also be symptoms of blight, but similar spots may be caused by rubbing or insects which do not pose a significant problem. Where you see the symptoms of blight, it is too late to control it. The affected nuts are likely to abort if the lesion appears before January and the affected shoots may be permanently damaged.

Walnut blight ecology

During winter, the main habitat for blight bacteria is in the dormant buds. Walnut researchers believe that the population is constant during winter because there is no available food for the population to expand and the temperature is too low. Even so, a large number of bacteria are waiting to reproduce when conditions improve in the spring.

Once the buds burst and growth shoots develop, there is potential for the blight population to expand. The optimal temperature for blight bacteria is 28-32°C, but they reproduce and lesions form at temperatures above 12°C. Rain, mist, high humidity or heavy dews or spray from irrigation help to create ideal conditions for blight. It is believed that when the population reaches a critical threshold, it becomes pathogenic and damages the host material.

The competition with other bacteria and natural resistance of the tree is not yet fully understood, but the researchers are continuing to look at ways of boosting the trees’ natural defences.  There is evidence that blight bacteria are able to enter wounds caused by wind or frost.

Blight causes little further loss of yield beyond about the first of February. Nuts that become affected when quite mature with a well-formed shell are much less likely to abort or become unacceptable for sale, as the blight is unlikely to penetrate into the kernel. The generally dry, low humidity conditions that tend to prevail in late summer are also less favourable for blight.  In addition, the blight bacteria have a natural ‘predator’ that exists in the tree canopies – a bacteriophage that is specific to the walnut blight bacteria.  The bacteriophage population builds up during the season, in parallel to the bacteria populations, so tend to be more numerous by late summer.

Blight control in the orchard

Growers rely primarily on copper-based sprays for blight control in the orchard.  Copper ions are toxic to blight bacteria, but such sprays also kill other harmless bacteria that could compete with the blight bacterium. 

To kill bacteria, copper ions must contact them, so the spray should penetrate the bacterium’s habitat as much as possible.  It is generally accepted that during winter, sprays cannot penetrate the dormant buds and spraying is not effective.  Fortunately, the bacteria are not causing damage at that time.

Spray timing is critical. Our research has shown that once the bracts pull back on the buds in spring – typically mid to late September in Mid and North Canterbury – you should begin your spraying programme. The first spring spray at bud burst is the most important of the season. (See the coloured page ‘From Bud to Leaf’ for pictures of buds bursting.)

Bud development varies considerably from location to location and even within a branch on a single tree. On older trees that aren’t being pruned hard, it’s easiest to look at the terminal buds, and spray when around 50% of these are 10 to 15mm long.  For young trees that have had some terminal buds pruned off, look at the buds near the ends of branches and leader and spray when around 80% of these have their bracts pulling back. The spray is effective at bud burst because it knocks the population while it is still small and is mostly in exterior tissues that can be reached by the spray.  If young trees are damaged badly by frost and re-shoot, it is a good idea to spray at this time too.

The latest research results suggest that Mankocide, a product combining both Kocide (copper hydroxide) and Mancozeb, is the most effective spray treatment for blight. It was significantly more effective than any of the other sprays tested. Incidentally, it was much more effective than either Kocide or Mancozeb applied alone, so it appears that the two products act together.

There’s a similar synergy between Mancozeb and the copper ammonium acetate product Liquicop, and it suggests that we should be able to reduce the proportion of copper-based products such as Kocide and Liquicop by combining it with the Mancozeb. Reducing copper has some real environmental benefits and also lessens the prospect of blight becoming resistant to copper. (See the research section Hot Off the Press for more information about the research on blight.)

If you are growing organically, you won’t be able to use Mankocide (or Mancozeb). An alternative is a Bordeaux mixture of 1:1:100 (1 kg of copper sulphate to 1 kg of hydrated lime to 100 ml of water).

Our Australian research colleagues recommend spraying every 7-10 days for the first 3-4 applications, with further applications every 7-14 days depending on how often it rains. They recommend continuing to spray until the walnut shells harden. Bear in mind that the Australian researchers were working in North West Tasmania where the rainfall can be more than three times as high as in Canterbury or Otago.

In New Zealand, most growers have more conservative strategies, but they vary – including every 2 weeks, or every 4 weeks, or before or after rain. If you are basing your strategy on rainfall events you should spray no later than 12 hours after rain. The spray will remain on the buds and continue to kill the bacteria until it is washed off by rain.  You will also need to keep an eye on rapid spring shoot growth, and spray to protect this new growth.

Whatever your strategy, you can stop spraying by February 1, or earlier in a dry, low humidity season when there’s little evidence of blight.

See the section ‘Hot off the Press’ for more about the research on blight and other research topics.

4.2  Other diseases to watch for

We are fortunate that we do not have the range of walnut diseases common in other countries, but we do have some.

Fungi from the Phytophthora genus and the Armillaria genus are known to cause significant root and crown rot problems.  The visible symptoms above ground are defoliation, loss of shoot growth and yield and eventually death of the tree.  If the infection is severe the tree will die very quickly, lesser infections cause death in two to five years.

In California, species of the Phytophthora genus have been most problematic.  In New Zealand, no species of the Armillaria genus have been known to cause infection in walnuts and all known instances of root rot are suspected to have been caused by a species from the Phytophthora genus, cinnamomi.

The fungi multiply rapidly in saturated soil. Make sure your walnuts are in free-draining soils and where the water table does not rise into the root zone. Avoid excessive irrigation, contaminating the site with soil-borne fungi on machinery, or flooding from run-off water from neighbouring sites – especially if they are contaminated. If you need to plant or replace walnuts in a contaminated site you can reduce the fungi by fumigation with Ridimol. Aliettte and Foli-R-Fos also work. Follow the instructions carefully and continue to be careful to avoid saturating the soil.

In New Zealand we have seen a few examples of gall, shallow bark canker and at least one North Island orchard has some infection from Armillaria.

Codling moth (Cydia pomonella) is certainly present in New Zealand, and is a problem in apple orchards, but growers are not finding significant infestations in walnut orchards.  Since there is no biological control available for codling moth, California growers have to manage it with insecticides that also kill natural predators of many other insect pests. The pests too, become a problem and have to be managed with insecticide.

CHAPTER 5 – Certified organic management

5.1 Introduction

You may decide simply to grow ‘as organically as possible’ for your own satisfaction, but certification is necessary before the Commerce Commission will allow you to sell your product as organic and for obtaining a price premium for certified organic product.

Consumers interested in organic walnuts expect a certified product. You can apply for full registration under a scheme such as Bio Gro New Zealand, Certenz, Demeter, Hortenz or Organz.  Once you have been through the transition period and have obtained full certification, you will be able to market your crop with the scheme’s registered trademark.

It is probably fair to say that most growers, across most horticultural and orchard crops, have a general preference for using the minimum toxic sprays possible while maintaining the quality and yield of their produce.  Going for full certification is not as difficult as many people believe.  One organic grower says the real challenge is in the mind, not the practicality of doing it.  The certifying bodies will outline simple steps for you to follow.

The environmental aims underlying organic production are:

  • To reduce the use of fertilisers, sprays and other compounds that are produced artificially using processes with high embodied energy or non-renewable resources.
  • To produce a product that is free of non-natural materials, in response to the wishes of those consumers who believe it is a safer or healthier product.
  • To avoid contaminating the growing environment with artificially-produced or toxic compounds, so reducing potential hazards to humans and other organisms.
  • To maintain a naturally functioning ecosystem of pests/predators, plants, insects, soil microbes etc., within a commercial productive system.


5.2 Sprays and organic practices

Compared to many soft fruit and vegetable crops, walnuts are relatively amenable to organic orchard systems.  Even in a conventional orchard system, insecticide and fungicide sprays are very seldom used for walnuts in New Zealand.  Copper spray is used for control of walnut blight, and it is allowed in an organic system in restricted amounts.  The problem with copper is that, after several decades of unrestricted use, it can build up to excess in the soil and discourage soil microbes and worms.As an organic grower you would need to be very careful about your timing, rates and methods, so that you achieve the maximum effect from minimum spray.

Heavy mulching is the main method of weed control in the drip zone, and it takes more time and effort than spraying.  A mulching mower, while expensive, is a valuable tool in that it mows and throws the clippings into the tree rows in a single pass.

Organic walnuts are a valuable niche product.  Currently at least one processor is paying a premium.  The demand far exceeds the supply because so few growers have chosen the option of a developing a fully certified organic orchard.

CHAPTER 6 – Shaping your walnut trees

Aiming for growth

Tree training is an essential part of the management in the early years of a walnut orchard. It encourages trees to favour vegetative growth against reproductive growth – which is important in quickly achieving full canopy cover in your orchard (unless you initially plant your trees at very close spacing). We are working toward a strong branch framework that won’t break in the wind.

We also want to allow the sun to have maximum access to the canopy of the mature trees – not just the upper leaves. For walnuts, that often means reshaping the tree.

Pruning the lower branches will allow you to work along the rows with your tractor and spraying equipment. And it has another benefit: eventually, you will have the option of cutting the trees to harvest valuable timber.

After about six years, your training should be complete, and you can allow cropping to take precedence.  From now on, you will simply be pruning to maintain your trees’ health and to allow the light to penetrate.

In areas where frosts have ceased by late September, you can do all your training and pruning in the winter when the trees are dormant.  Canterbury growers can do most of their training in the winter – except very young trees, which should be trained after the last frosts.  In Otago, growers generally do all their training and pruning in the summer.

A variety of views

Experienced growers can have differing views on training, but they agree on the principles. The differences are to do with the appropriate techniques for individual sites and NZWIG’s research committee is organising a trial to test various strategies on light, medium and heavy land.

Ross Jamieson takes a conservative approach on lighter land in the early years and advises growers not to create a ‘palm tree’ look. He prefers to maintain as much of the canopy as possible and, referring to the article you are about to see, he says: ‘Mid-branch pruning will only promote a growth response if the health of the tree, the management and the environment permit it.’

‘It’s a good option for fertile soils’, Ross says, ‘but on the poorer soils and in harder environments it’s better to cut back less severely. Continual harsh pruning in hard environments will reduce growth because the tree will need to replace the canopy by using stored nutrients. With a gentler approach the leader and branches will usually be the same length at the end of the season as they would if you had cut them back severely because there’s less length to make up.’ Ross warns that heavy pruning can reduce your crops in the early years.

David Murdoch has advised many growers on training and pruning. His article suggests two approaches – a ‘modified central leader’ technique and a more conservative variant ‘a minimum pruning option’. On fertile soil with good shelter, you can use the modified central leader technique. On lighter soils, you may prefer the ‘minimum pruning’ option (or even be more conservative still).  In any case, you will probably use these instructions as a starting point, and will refine them over time to suit your orchard’s environment, as you observe the results of your training efforts.

These instructions take you through walnut training year by year. Note that appropriate training for a 4 year old tree is not the same as that for a one year old tree!  As David Murdoch shows, the training cuts evolve as the tree grows. In the first few years, you will be aiming for trunk height, then in the next few, for development of strong, well-spaced branches.

6.1  Training young walnut trees

By David Murdoch, Peninsula Tree Nursery


Before considering a training system for your young walnut trees you need to consider the following:

  • Which cultivars will you grow?
  • What planting densities are right for you?
  • Will you grow your walnuts as a hedgerow?


These factors will largely determine the training system you will use.The main options are:
• No training
• Vase system
• Centre leader
• Modified centre leader
• Hedgerow

Within the local industry at present quite different views are held as to the best or most appropriate system. Most seem to agree that the no training and the vase shape are not suited to New Zealand conditions due to the limb damage caused by the wind loading we have in most growing areas (see Figure 6.1). The most suitable option from the rest will, to a large extent, depend on the planting density and the individual’s preference. In my view, the modified centre leader system is the most appropriate as it encourages large vigorous trees with well-spaced scaffold limbs that are less prone to wind damage. Given that the common planting density is still 10 metres by 10 metres, these trees would reach a reasonable size and shape before nut production at six to seven years of age. It is also the most commonly used system in the USA and Europe.
Why not the centre leader system used for fruit trees and in forestry?
Apples are grown on dwarfing rootstock and the weight of the fruit is used in the training process. Neither of these features is available for the walnut grower. Dwarfing rootstock controls the growth rate and, to a large extent, the height of the apple tree, so tree size is much smaller than with walnuts. The weight of the fruit is used to lower the branch angles, exposing them to more light, which initiates more flowering/fruiting/weight. These fruiting scaffolds are replaced on a regular basis.

In forestry, initial high density planting forces trees to grow up to compete for light.  Tree provenances have been selected for upright growth.  Again neither of these factors are available to the walnut grower.  The centre leader system does have a place in walnut orchards as it is the system most suited to high density hedgerow plantings (we will look hedgerows later).  However, while they can be forced into it, grafted walnuts are not natural centre leader trees (see Figure 6.2).

The terminal buds flower while the trees are quite young which encourages vegetative buds lower down on the stem to shoot.  The presence of the terminal fruiting shoot forces the lower, more vigorous, vegetative shoot off to one side so the new leader becomes off-centre hence the terms, modified, or off set, centre leader.  Late spring frost will also cause this type of growth.  Therefore the modified centre leader best emulates the natural form of grafted walnuts and is the most suitable system for medium to low density orchards.

6.2  Principles of tree training

There are many reasons for training trees.  Commercial growers need to grow their trees to a specific shape to maximize production within the required specifications for a given crop, to allow vehicle movement through the orchard, and to develop a strong framework less vulnerable to wind damage etc.  While shape is less critical in the home garden or small orchard there are advantages in correctly shaping your trees.  Let’s look at some basic principles that should help you make the right pruning decisions.

Responses to pruning

Trees growing under good light conditions with adequate supplies of water and nutrients grow in proportion to the amount of active leaf area.  The leaf is essentially a solar panel designed to capture sunlight and through photosynthesis convert this light energy to carbohydrates (sugars) which can be transported around the tree.  Therefore any removal of leaf material through pruning will have a negative impact on the overall growth of the tree.  However, since the removal of material reduces the number of buds remaining on the tree there are more carbohydrate reserves available for these buds.  It is basically a balancing act between the tree’s reserves, its root system, and the number of buds – the growing points.

The tree grows for a season producing a certain amount of shoot and root growth; pruning removes shoot material but not root material.  Therefore, the extra capacity of the root system is directed to fewer buds resulting in increased length of those shoots.  By the end of the growing season the roots and shoots are in balance again so the increase in vigour is not maintained the following season unless you do some more pruning.  It is this invigorating effect that you can be use to train young walnuts to the desired form.


  • A good pair of secateurs, preferably roll cut
  • A good pair of loppers, once again preferably roll cut
  • A pole pruner for high pruning
  • A sharpening stone and/or a fine file
  • A pruning saw
  • A pruning ladder

Training should begin when the trees are young and generally should only require secateurs. If you have to resort to the pruning saw or worse, a chainsaw, then you have really left it too late and are trying to correct past mistakes.

Pruning Cuts

Heading – a heading cut is where a shoot, usually one or two year old, is cut back by a third or more to force a vigorous growth response (often used to develop a strong leader) or to encourage a branch to develop in a selected direction.

Thinning – a thinning cut is used to remove a shoot, where it arises from either the trunk or a main branch, to reduce over-crowding within the tree.

Pruning Cuts

Heading A heading cut is where a shoot, usually one or two year old, is cut back by a third or more to force a vigorous growth response (often used to develop a strong leader) or to encourage a branch to develop in a selected direction.

Thinning A thinning cut is used to remove a shoot, where it arises from either the trunk or a main branch, to reduce over-crowding within the tree.

Where to prune

Make a heading cut just above the bud and on an angle away from it.  The angled cut directs water away from the bud. Moisture lying in the crevice between the bud and the stem provides ideal conditions for walnut blight.

A thinning cut is made as close to the stem collar as possible without cutting into it.  The collar around the base of the stem contains the healing cells.  These will swell up and roll over the cut surface eventually closing off the wound.  The worst thing you can do is to leave a stub as this stops the callus from rolling over the cut and acts as a source of food for fungi which may then grow into the tree.  Likewise cutting too close to the trunk damages the collar slowing down the healing process.  It can stop the callus cells from rolling over the wound altogether.  The result can be a wound open to invasion from pathogens.

Pruning paste has little or no effect on the healing of a wound – in fact it can help rot to spread.

When to prune

Winter is the most usual time to prune trees.  With no leaves, it is easier to determine the existing shape of the tree and make the desired adjustments. Winter pruning has a less debilitating effect because no leaves are being removed so carbohydrate production is not affected.  However, since the tree has stores (from the previous growing season) of nutrients and carbohydrates in the stems and roots these become available to the remaining buds promoting a vigorous growth response.

Spring pruning is somewhat problematical with walnuts.  Walnuts bleed profusely when pruned in spring; but on the other hand, if pruned in late winter they are vulnerable to damage from spring frosts.  It is of particular concern with very young trees, but in some areas also with older trees.  There is no real answer to this problem and it will depend to a large extent on the particular location and the cultivars being grown.  Generally it should be safe to prune trees older than three years in late winter or early spring.  The weather conditions at the time will influence the amount of bleeding.  Try to choose a period when the weather is dull and avoid a period with frosts followed by sunny days because sap flow is particularly strong under those conditions.

For one or two year-old trees which you have to cut back hard, late winter/early spring is too risky because frost can devastate them.  In frost prone areas you will need to delay heading back until early to mid-November.  Bleeding will be a problem and to a large extent we have to live with it.  However, choosing cool weather conditions will help.  The actual debilitating effect is probably not significant.  The carbohydrate and nutrient reserves contained in the stem and root system are directed to the reduced number of buds remaining on the young tree resulting in a very strong growth response from the heading back.

Summer pruning is often used for thinning unwanted shoots. The growth response from summer pruning is less vigorous than from winter pruning so it is useful for slowing down the growth of vigorous trees. However, avoid excessive pruning at this time as it can be quite debilitating. Remember that the tree has expended reserves to develop spring growth and will not recover this back from carbohydrate production by the leaves for quite some time.

What to prune

What you cut will depend on the form you want the tree to have, e.g. modified centre leader, centre leader, vase, etc.  Each system is generally a combination of heading and thinning cuts.  However, there are pruning cuts you should make regardless of the tree form:

  • Remove any suckers arising from the tree base.
  • Thin any forked leaders back to one shoot, this applies to forked branches as well.
  • Remove any damaged shoots.If these occur on leaders or critical branches head back below the damaged area and during the growing season, develop a new leader from the best re-growth shoot and thin out all the others.
  • Thin out any crossed or crowded branches.


6.3 The modified centre leader system

The modified centre leader system for walnuts is reasonably straightforward to develop. However, it does require some courage from the grower because a considerable amount of hard pruning is required to achieve the desired shape.  This can be quite nerve wracking for the new grower and the concept of hard pruning to stimulate a vigorous growth response is often difficult to understand.

The aim is to create four or five well spaced (30 to 50 cm apart) scaffold branches and an off-set leader which will become the fifth branch.  The strategy will form a round to oval headed tree, similar to the natural shape of walnut trees, which will adequately fill its allocated space in medium density orchards.  The training generally takes 5 to 7 years and should be completed before the trees start coming into production.  The broad principles of this training system are:

The development, in the first two years, of a strong single shoot at least 1.8 to 2 metres tall (Photograph 2).

Prune the trees to a single shoot during the first growing season.  The following spring, head the shoot back to five or six buds above the graft, a month or so later select the strongest shoot for the new leader and pinch the others back.  Given reasonable growing conditions the shoots should be 1.8 metres plus by the end of the season.

There are other options, such as allowing the leader from the first season to grow on while pruning off all the side shoots that grow during the season.  In my experience this tends to develop thin trunks prone to wind damage, the tree becomes top heavy and blows over. Problems also arise if the terminal buds flower and vegetative growth stops for an extended period.  If the buds don’t flower (which can depend on the cultivar) then you may prefer this method.  If so, make sure that you leave two or three temporary branches on the main shoot in order to feed the trunk and thicken it.  Personally I find it much more straightforward to head back than fiddle around with the alternative methods.  Remember that hard pruning generally results in a vigorous growth response and growers who do it will tell you that it does work.  You will get runts, but that is usually due to other problems and the odd plant that can’t stand the stress and will give up totally.

The heading back of the leader the following spring

The leader should be headed back by about a third or between waist and chest height depending on the overall height of the tree (Photograph 4).

The top third of the shoot is usually quite pithy inside.  This wood has a fluted appearance on the outside.  As you move down the shoot the wood becomes more rounded and is less pithy inside.  To ensure a vigorous growth response it is important to cut back to this round wood, regardless of the tree height.  Cutting back to pithy wood will give a poor growth response.  Likewise never cut back to a neck bud.

The selection of the first scaffold (framework) branch and removal of all other shoots during the following winter


When selecting a shoot as a branch make sure you choose one that is growing on a flattish angle (between 90 and 45 degrees angle to the trunk), never choose one sharper than 45 degrees as they are generally weak and suffer from bark inclusion (see diagram 1 above).  Neck buds (see Photograph 1 above) develop into sharp angled branches.

The leader is headed back to encourage another breakout of shoots higher up the tree (see Photograph 5).  The amount the shoots are headed back can vary depending on the growth response required. The harder they are headed back the more vigorous the growth response.  However, sometimes the new shoots grow too strongly and are vulnerable to twisting and bending down before the bark has lignified – particularly after strong winds.  In more exposed orchards a lighter heading back would be more appropriate although the growth response will be less.  Always make sure the leader, after heading back, is well above the next highest branch and it has been headed back hard enough to ensure the growth response is strong enough to compete with the regrowth from the branches.  If you are heading back branches always head back the leader.  Failure to do so will result in a branch overtaking the leader and you will lose control of the training process.

To allow you to move machinery in the orchard, the first branch should be at least 1.8 metres above the ground.  For various reasons the branch may only be about 1.5 metres or less; if so it must be treated as a temporary branch to be removed within two years.  You could make an exception for those aligned in the row where you won’t be using machinery.

Training continues on the same pattern in the second year

By the following winter the tree will have grown a new leader and several strong branches in response to the previous pruning.  (See Photographs 6 and 7)

  1. Head the leader back by a third or more.
  2. Select the second branch and head it back.
  3. Head back the first branch (selected last year) but not as hard as the other branch or leader.

Depending on the growth the tree has made, it is sometimes possible to select a third branch at this stage, just make sure you have at least a 25 to 30 cm gap between them.  If the first branch selected last season is too low, now is a good time to remove it before it becomes too strong.  You will notice that a headed branch responds in much the same way as a headed leader i.e. a breakout of a number of strong sub-branches.

The scaffold branches should be treated in the same way as the leader.  In other words, where two sub-branches are growing opposite each other, thin one of them out.  You should aim for a gap of at least 20 cm between sub-branches, slightly less than for scaffold branches on the main trunk.  If one of the sub-branches is stronger or growing in a more suitable direction than the branch leader, it is fine to prune back to the sub-branch and use it as the new branch leader.  You can, to a certain extent, control the direction the branch leader grows in by heading back to a selected bud – i.e. you can send it left or right, up or down by selecting the appropriate bud.  Remember though that trees have their own sense of purpose and you can only manipulate them to a certain extent.

Selection of the final scaffold branch and permanent leader by year five or six (Photograph 8)

The timing will depend on the individual tree’s performance.  You may be able to select two or even three scaffold branches in one season.  It’s just a matter of whether or not you have enough well-spaced branches to choose from and that can vary from tree to tree.  If you don’t have enough branches, head back the leader as usual and select the final branches next year.  If you are happy with what you have, then in years five and six training really consists of some final fine-tuning.  Head back the leader a little so it retains its vigour and remains dominant over the scaffold branches.  If required, redefine the direction of a scaffold branch, thin out any damaged or crossed over sub-branches.  Remove that first scaffold branch if it is too low and any other shoots that may get in the way of orchard floor management.  Any forked branches should be thinned back to one shoot.

6.4 Modified centre leader – minimum pruning option

A variation to the standard modified leader is the minimum, or reduced, pruning option.  This system follows the methods outlined for the modified leader system, i.e. heading back the selected leader and scaffold branches.  However, rather than pruning off the surplus growth most of this is retained for nut production.  The system has both advantages and disadvantages:


  • Some early nut production particularly with precocious cultivars such as Dublin’s Glory (W/1S/143) and, to a lesser extent, Rex (W/1C/152).
  • A general reduction in tree vigour.It would suit moderate planting densities e.g. 10m by 10m with one in the middle or 7m by 7m etc.
  • It is less stressful for the grower!It is often difficult for the new grower to understand the pruning required with the modified centre leader system.



  • A more complex system. With the modified centre system the leader and scaffold branches are selected and everything else is simply pruned off.The reduced pruning system requires more thought and time as decisions need to be made about what to leave on.
  • A slight reduction in vigour, which may be of some significance in orchards with wider plant spacing.


Years one, two, and three

The training in these years should follow the pattern set out for the modified centre leader system already outlined.  The modified centre leader system stimulates early tree vigour, which is essential for building the tree’s framework.

Years four and five

Reduce the pruning in year four.  The leader and permanent scaffold branches are headed back so they retain their vigour, any strong branches competing with the leader or a scaffold branch are thinned out.  All other branches are left to produce nuts the following season.  Photo 10 is an example of a five year-old tree prior to pruning.  The leader and scaffold branches have been selected and will be headed where illustrated.  Four branches have been marked for thinning.  One is competing with the leader and must be removed.  The other three are part of a group of four that are too close together.  They should be removed to prevent overcrowding and to maintain the dominance of the leader. All the other branches can be left for nut production.  However, they are not permanent branches and will have to be thinned out over time.

6.5 When things go wrong with the training

Trees are subject to the vagaries of the weather, pests and diseases, accidents and poor management.  The object of any training system is to guide the tree in the direction you would like it to go but bear in mind that tree training is not an exact science and that every tree will be different.  The training system has to be a flexible set of rules which may vary from tree to tree.  For example, one tree may have two branches suitable for the second and third scaffolds but in theory they are too close together.  It would be pragmatic to accept this, providing there is a reasonable gap between them, rather than force a less suitable branch into a scaffold.  The next tree in the row may have a large gap between the second and third scaffold (a common occurrence with Meyric). Again the best option is to accept the spacing and move on.

The greatest influence on growth of young walnuts is the weather pattern for a particular season.  Walnuts thrive on heat so growth is better when spring and summers are hot.  There is not a lot you can do about it; just continue with the training and accept that in some years the response to your pruning will be better than others.

Late spring frosts can cause severe damage, particularly if they happen in mid to late November, because the initial spring growth flush can be burnt off.  In areas subject to late frost you can delay heading back your young trees until mid to late November.  However, do head back as the vigorous growth response will help your trees to grow up out of the frost zone quicker than not pruning at all.  On three or four year and older trees, prune in the winter regardless; they will make up any growth lost to late frost.  However, you will need to do some corrective pruning after the frosts to ensure the headed leader and branches maintain a dominant lead shoot.  Frost will not kill older walnuts as they carry many dormant buds that will shoot and replace any damaged spring growth.

Sometimes only the leaves and terminal buds of the spring growth are burnt and new shoots grow from the cluster of sub-terminal buds.  It often leads to stag head growth where two or more new shoots, in close proximity, grow with more or less equal vigour.  Allow all the shoots to continue growing until they are 10 to 20 cm long, then select the best shoot growing more or less in the right direction and either pinch the tips out or prune off the others.  Pruning off (particularly on the leader) is generally the best option as the pinched shoots sometimes send out strong lateral shoots later in the season which can compete with the selected shoot as well as causing confusion and crowding.

Sometimes one or more trees may show a consistent lack of vigour while others next, or near, to them grow well.  There can be several causes for the runts.  Firstly check the tree carefully to see if there is any damage or disease to the trunk or around the root collar which may be restricting growth. Check the graft union to make sure the grafting tape has been removed and for any abnormalities.  While you can’t check the roots without digging up the tree, look for signs of yellowing, or other abnormalities, on the leaves that can indicate root system problems.  If everything looks fine, try hard-pruning the tree to stimulate a growth response.  If that doesn’t work, try moving it to a different spot e.g. the vege garden for a year or two.  If the tree is damaged or appears to have root problems, the best option is to replace it with a new one.

If most of the trees in the orchard are showing a general lack of vigour, you will need to review your management practices or improve your shelter.

The growth response from pruning on walnuts is very strong. However, the bark on young shoots is very soft and tender until it liginifies, by which time the shoot may be a metre or more long.  These shoots are vulnerable to twisting and bending under windy conditions which can split, or otherwise damage, the bark near the base of the shoot.  This problem usually occurs only on branches that have been headed back very hard, which is sometimes necessary, so the new growth is very vigorous.  These shoots can be tied to a bamboo stake and braced back to a point higher up the trunk with string until the bark lignifies.  If the problem occurs often, avoid hard pruning, only head back by about a third of the branch length. If you do have to cut back hard to improve branch angle or direction, be prepared to use the bamboo stake for support.

Training walnuts is not difficult but it does take some practice.  You will make mistakes but as you become more experienced each year you will become more confident and be able to correct mistakes.  The most difficult part is the initial pruning i.e. overcoming the nervousness of cutting back expensive young trees that appear to be growing quite happily.  The trees are going to be there for a long time and sensible training when they are young will avoid some expensive limb damage when they are older.

6.6 High density plantings

So far we’ve focused on the modified leader system because most orchards in N.Z. are planted at the wider spacing of 10 metres by 10 metres.  However, closer spacing is more popular overseas and should be more widely used here – particularly on the lighter soils where growth can be controlled more easily – by manipulating the irrigation.  The actual planting distance is still a point for discussion and to a large extent will depend on the soil type, cultivar, and your preference.  Walnuts, unlike most fruit trees, will grow into quite large trees and we currently lack less vigorous or dwarfing rootstocks with which to restrict growth.  The use of root control bags has been tried with some degree of success but cost is still a factor.  However, if you have a small area of land, the bags would be well worth considering.

Generally the distances between plants in the row should be no more than four metres apart and no closer than three metres. The distance between rows will depend on the size of the orchard machinery and equipment you want to use. A large orchard will require larger machinery and you would need a greater gap between rows for access.


The choice of cultivars is important for high density planting.  The system is only suitable for less vigorous more precocious lateral bearing ones such as Rex (w/1c/152), Dublins Glory (w/1s/143), Tehama etc.  Others such as Meyric (w/11h/1199-4) and Stan (w/Ble 300) are not suitable because of their vigour and less lateral bearing habit.  If possible, less vigorous rootstocks should be used for the planting stock.


The object with high density planting is to create a hedgerow and form a fruiting wall where each row becomes a unit and individual trees are less important.  In a lower density orchard each tree becomes a fruiting canopy.  To restrain the size of the hedgerow and keep up the production, you would need to use mechanical hedge trimmers.  It is not a ‘plant the trees and do nothing’ system.  Walnuts require light to initiate flower buds.  Because of the closer row spacing with high density planting the fruiting wall should be trimmed at a suitable angle.  Diagram 3 illustrates the most suitable angle for macadamias in Australia.  It would probably be appropriate for walnuts in New Zealand.


  1. For the first year or two treat the tree as you would for the modified leader system described earlier – i.e. develop a single leader 1.5 to two metres tall.
  2. Head back this leader to round wood to encourage a breakout of branches.
  3. Select the most upright shoot to form the new leader and head it back by about a third.Select the first branch approximately 75 cm to 1 metre above ground level and head this back, making sure it is at neither too sharp or too flat an angle and it is growing along the row.  Prune all other shoots off.  The trunk on hedgerow trees is much shorter than on other systems but should be at least 75 cm to allow for weed control, mowers and the option of using tree shakers in the future.
  4. Several shoots will have grown from the previous heading back.You should be able to select two of these on opposite sides of the tree and growing along the row.  Head back these shoots and the leader.  Remove any other vigorous shoots growing out into the row and any others growing at too sharp an angle.  Leave all other shoots on unless they are growing on the trunk below 75 cm.  The headed back shoots will form the basic framework of the tree.  The gap between them is less critical than with the modified leader system, 15 to 20 cm is fine.

Continue the pattern outlined in (4), selecting more framework branches and heading these as well as the

leader back. Remove any very strong shoots growing out into the row. You want to aim for moderately vigorous shoots carrying lateral fruiting branches because they will form the fruiting wall along the row.

  1. Depending on the size and vigour of the trees, you can now start using a mechanical hedge trimmer. Set the blades and angle so that no more than a third of the new growth is removed. Too hard a trim will reduce nut production while trimming too lightly will reduce shoot vigour and result in thin twiggy growth.

Points to remember

  • Always prune hard enough to maintain vigour
  • When heading back branches cut back to a bud that will grow more or less along the tree row.You may need to summer prune to correct the shoot direction.
  • Always maintain the leader as the dominant shoot so continue to head it back for at least five to six years.
  • We have very little experience with high density walnut orchards in New Zealand so growers will have to experiment, to a large extent, when developing such an orchard.


6.7  Additional comments from other growers

‘With grafted trees, the terminal bud at the end of a season’s growth shoot has a high likelihood of fruiting the following year.  So to achieve maximum vegetative growth, head the shoots back to stop them fruiting.’

‘Many growers choose to remove large, low branches.  In California, the height of the first branch is typically 1.2 metres above the ground – high enough to permit access for orchardised mowers and tree shakers.  In New Zealand, some believe that the timber grown in the butt log of the trunk will be of significant value (if properly pruned) and therefore choose a first branch height of two or even three metres.’

‘If you cut any unhealthy wood, sanitise the cutting blades before you reuse them – to avoid the risk of spreading the disease. Seventy per cent ethanol is generally regarded as a good sanitiser but methylated spirits is an acceptable substitute.  Ensure that all of the sanitiser has evaporated from the blades before you re-use them.’

6.8 Pruning

With trees more than 7 years old, you will prune rather than train.

Mature trees are pruned to maintain light penetration and yield, and to remove any diseased or dead wood.  It is difficult to advise what and how much pruning we will need to do, because there are few mature walnut orchards in New Zealand.  In comparison to other crop trees, it is likely that pruning needs will not be high in walnut trees. The nuts fall to the ground rather than being picked, so allowing the tree to become large will not hinder our harvesting.

It is generally not necessary or beneficial to prune a walnut tree to force shoot growth to provide fruiting sites because the tree is able to produce sufficient fruiting sites without pruning.  Certainly, it is best to remove dead wood and prune to expose the canopy to light.  In hotter climates such as California, where trees have much more vegetative growth, it is necessary to prune heavily to avoid overcrowding of trees (which causes reductions in the canopy volume with sufficient light to fruit).  It is likely that overcrowding will occur more gradually in the New Zealand climate.  Even so, it may eventually be necessary to remove whole trees (if they have initially been planted at close spacings) to maintain light penetration.

If your trees have not been correctly trained when young, you may have to make large pruning cuts to remove weak or crowded branches.

CHAPTER 7 – Using your irrigation

Walnuts require adequate moisture for commercial production.  They are intolerant of the waterlogging which is mostly associated with high water tables or impeded drainage.  Your soils must be free draining and ideally have good moisture-holding capacity.  In many eastern areas of New Zealand the free-draining soils are often shallow with stony subsoils and allow the moisture to drain away too easily.  If you are growing walnuts on such soils you will need irrigation to make your production economic.

7.1 Soil Moisture

Soil moisture is the most important component in plant growth.  Moisture must be available to meet the trees’ evapo-transpiration needs.  Their needs will be on average 4-5 mm/day during summer, but up to 8 mm/day in a hot, dry wind.  The trees use very little of the moisture for cell production – most is transpired to help the tree to remain cool, just like our perspiration.  Soil moisture also has other functions:

  • as a solvent of nutrients that are essential for plant growth.
  • to help control soil air and soil temperature.

Growers are most concerned with the effect soil properties have on:

  • the moisture-holding capacity of soils.
  • the availability of the moisture to walnuts.
  • the movement of water into and through the soil.

Each of the factors is related to the size and distribution of the soil pores and the attraction of the soil particles for moisture.  The soil is a delicate balance of soil particles, air and moisture as illustrated in Figure 7.1.

As water is applied to the soil, either by rain or irrigation, air is displaced and the soil ‘wets up’.  Water fills the soil pores and may eventually replace all the air.  When all the pores are filled with water the soil is saturated (Figure 7.1, A).  The soil is now unsuitable for growth of walnuts (as is the case for most other plants).  If the supply of water continues, the soil is maintained in the saturated state and water continues to move rapidly through the soil.  Nutrients are leached through the topsoil and into the subsoil or groundwater.  If the supply of water is cut off, water will continue to drain through the soil.

After 1–2 days downward movement of water essentially stops.  The interval can be longer in heavier textured (clay) soils and shorter in light textured (sandy) soils.  The soil is then considered to be at field capacity (Figure 7.1, B).  At field capacity, water has moved out of the macropores to be replaced by air, but the soil micropores are still full of water.  In other words, the soil is as full as it can be without being saturated.  The water remaining at field capacity is held in micropores around soil particles by suction.  This is the source of water for plant growth and plants must overcome the suction to obtain water for growth and transpiration.  While water will continue to drain downward, the rate of movement is very slow.

Plants will remove moisture from the soil until the wilting point is approached.  The amount of moisture between the field capacity and wilting point is known as available moisture, as shown in Figure 7.2.  Initially the water is readily available to the plant, but as the soil dries out it becomes much harder for the plant to extract the water.  Different plants have differing abilities to overcome the suction binding the moisture to the soil particles.  Walnuts can extract about 60% of the available moisture and this is represented in Figure 7.1, C.  Once the plant has used the moisture between field capacity (the broken line) and the ‘trigger’ or refill point (the solid line), the readily available moisture has been depleted.

As the soil dries below this level toward the wilting point (the area shown as survival water in Figure 7.2) the walnut trees will show visible signs of limited moisture availability or moisture stress.  The leaves will be wilted during the daytime, especially on hot and/or windy days.  At night there will be renewed plant vigour.  During periods of moisture stress there may be some premature leaf loss, shoot growth may slow or stop, and the nuts may slow their rate of expansion.  When the wilting point is reached (Figure 7.1, D) the walnuts will be wilted both day and night.  The trees will die if they do not get water.  There is still a considerable amount of moisture in the soil at wilting point.  This moisture is usually found in the smallest micropores and is tightly bound around soil particles or is part of the clay mineral structure.  It is not available to the plant.

For a walnut tree, three forms of water are important.  Each has different implications for normal and continued growth.

Superfluous (free or drainage) water:

♦ soil moisture held above field capacity.

♦ loosely held in macropores at very low suction (0 to 20 centibars).

♦ undesirable and removed rapidly by drainage.

♦ leaches nutrients as it drains.

Available (or capillary) water:

♦ held between the field capacity and wilting point.

♦ held as films at suctions of 20 to 1500 centibars.

♦ some is readily available to walnuts.

♦ functions as a nutrient solvent.

Unavailable (or hygroscopic) water:

♦ held at very high suctions (greater than 1500 centibars).

♦ held in microscopic pores or part of the soil colloid structure.

♦ largely non liquid.

♦ not available to walnuts.

The amount of soil moisture available to walnuts is dependent on both the soil texture and the depth of soil.  Where gravels underlie the mineral soil and the depth of mineral soil is small (0–30 cm), the amount of available moisture is very low.  On these soils there is insufficient available moisture to sustain optimal growth without irrigation and, in most seasons, insufficient available moisture to keep the trees above wilting point.

Table 7.1 shows the average relationship between soil texture and available soil moisture (expressed in millimetres of water per metre of soil).  The soils with the best moisture holding characteristics are silt loams.

[1] To convert from mm/m (call this M) to volumetric moisture (call this V), use , e.g., 110 mm/m is equal to 11% volumetric.

Always consider the available moisture in conjunction with the depth of soil to gravels (if they are present in the sub-soil).  Gravels have very low amounts of available moisture, typically only 40–50 mm/m of which only 25–30 mm/m is readily available to walnuts.  Any soils with gravel or stones will have lower available water than a mineral soil of equivalent depth.  The trees can only readily extract a proportion of the available moisture.  For walnuts about 60% is readily available.

The available soil moisture capacity will determine the usefulness of the block of land for walnut production.  The available soil moisture will be the buffer between adverse climatic conditions like hot, drying winds, and economic production.

7.2 Irrigation Requirements

The amount of available soil moisture will also determine the need for, and amount of, irrigation.  For a deep-rooted plant such as the walnut, growing on deep soils with high available soil moisture, you may not need irrigation every summer.  If you are growing your trees on shallow and coarse-textured soils with low available moisture you will probably need irrigation every summer for a profitable yield.

You should consider these issues to assess how much irrigation you will need.

Establish the effective rooting depth of the walnuts.

The effective rooting depth is the depth of soil from which the significant proportion (75–85%) of moisture is extracted, as illustrated in Figure 7.3.  Clearly when the trees are young the rooting depth will be shallow, no more than 200–300 mm.  As the trees mature the rooting depth will increase.  Even though walnuts develop a deep root system, a mature tree may extract most of its moisture from quite shallow depths.  Even in deep, well-drained soils much of the moisture can be extracted from the top 700 to 900mm; i.e., the area of greatest root activity or the effective root zone.  This depth of soil will be less on shallow soils because there is less mineral soil, and may be as little as 200–300mm deep.  Pans, shallow water tables and gravel subsoils can reduce the effective root depth.

Determine the amount of readily available soil moisture in the effective root depth of soil

The readily available soil moisture is about 60% of the available soil moisture and is the reservoir to be used before irrigation is necessary.  Once the available soil moisture is depleted, and you will need irrigation to maintain productive growth, shoot extension and nut growth.

Use nearly all the readily available water before irrigating

Set an irrigation target so that slightly less than the readily available water is depleted.  The amount of readily available water below this target level is an emergency supply.  Once irrigation has begun, only apply the amount of moisture necessary to match what the trees are extracting from the soil – see Measuring Soil Moisture section for details on how to achieve this.

Demand for moisture depends on the age of trees, ground cover and prevailing weather. Young trees with smaller leaf area will require significantly less water than older trees.  Young trees will be extracting moisture from shallower depths because of the less developed root system.  They will require irrigation earlier than older trees because the smaller volume of available moisture will be exhausted sooner.  A grassed orchard will need more water than one that is bared or has a herbicide strip.  The grasses will be competing for the available water and increase the rate at which the reservoir of available soil moisture is depleted.

Don’t neglect the irrigation of your shelter trees.  Some species are sensitive to moisture stress and will defoliate if the supply of moisture is too low.

Take care to measure and record rainfall because it’s not only free but the most efficient source of water.  It replenishes soil moisture much more effectively than irrigation.  Good irrigation decisions and practices cannot be achieved if rainfall is not measured and recorded.

Walnuts are sensitive to soil moisture content and its availability.  Don’t use your irrigation haphazardly.

7.3 Measuring soil moisture

Monitoring soil moisture will help you to schedule your irrigation to achieve maximum growth of walnuts and shelterbelts, without wasting water or raising the risk of root rot.

You can contract an expert to do it for you, or carry it out yourself.  A contractor will install probe holes, measure some key attributes of your soil, and then return to your property every one to two weeks to make ongoing measurements.  You will receive a report detailing soil moisture levels and a recommended irrigation schedule so that you can bring the soil back up to a suitable moisture status.

The remainder of this section provides guidelines to help you set up soil moisture monitoring equipment for yourself, and interpret the results.  Before discussing soil moisture measuring equipment, we recommend that you get into the habit of recording rainfall and being aware of weekly E.T. (evapo-transpiration) readings for your district

Factors that influence soil moisture content

In theory, you can monitor soil moisture by calculating a water balance using rainfall, E.T. and other measurements.  Although the water balance approach is not currently a practical method for irrigation scheduling in New Zealand walnut orchards, the concept behind it is useful. Usually the soil is assumed to be at field capacity[2] at the beginning of the season.  Then, each week, you subtract the water lost from the soil during the week by evapo-transpiration and by drainage, and you add the water gained during the week from rainfall and irrigation, to obtain a new estimate of the soil moisture level.  Even though several of those factors are difficult to measure in practice it is worth being aware of all four factors that influence your soil moisture: E.T., drainage, rainfall and irrigation.  If you have a good feel for how they have varied over the past week, you are in a much stronger position to understand and interpret your soil moisture readings.

For example, if you know that E.T. has averaged 4 mm/day over the past week, and you have had no rain, then you might expect to add approximately 28 mm of water by irrigation to make up for the week’s E.T.

Technology and estimates

Some growers are quite relaxed about soil moisture measurement.  Their approach is to dig a hole once a week, feel a handful of the soil, put that information together with their rainfall reading and a knowledge of how much water their trees ‘usually’ need, and turn the irrigation on for a duration that seems right.  In general, they don’t go too far wrong, but may not be optimising the growth and production of their trees.  They also run the risk of over-watering, and the probable consequences of root rot.  The constant digging also disturbs the soil around the roots of their trees.  It’s worth considering the soil moisture measurement systems available.

If you have several different soil types on your property, you will probably need to install a soil moisture measurement probe in each type.  If you have been able to design your irrigation system so that the different soil types are in different irrigation blocks, each probe reading will help you prepare your irrigation schedule for one soil type. If you have more than one soil type within a single irrigation block, you will need to base your schedule on the average readings from the probes.

Remember that walnuts are deep-rooting when mature and ideally you would like to know the soil conditions down to a metre depth.  However, even for mature trees, much of the root activity is in the top 300mm or so (depending on your soil type).  Installing a vertical probe that spans between 100 mm and 300 mm depth provides you with sufficient information.

The ideal time for installing of moisture sensors is late autumn when the soil is damp and soft and it’s easy to install the sensors.  Importantly, the disturbed soil can settle around the sensors during winter, so that they can provide you with accurate readings from the end of winter onward.  The soil usually ‘fills’ with water during the autumn and winter, so the end of winter is the best time for determining the field capacity of your soil.  You need to know the field capacity for your irrigation scheduling.

[1] Most newspapers report the E.T. figures daily or weekly.  They are measured for a grass sward, but are a useful indicator for your orchard.

[1] When soil is at field capacity, it is holding the maximum amount of water possible in its micro-pores, but the water has drained out of its large macro-pores, leaving them filled with air.  Thus there is water available for plant growth, but the soil is not saturated.

Soil moisture probes

Soil moisture measurement probes come in two types:

  • Those that measure ‘tension’ – i.e., how tightly the water is bound to the soil, and therefore how hard the plant has to ‘suck’ to get it – the tension will be high if there is not much water in the soil and low if there is a lot.
  • Those that respond to the soil’s volumetric moisture content.[3] The neutron probe measures moisture content fairly directly by counting hydrogen ions, but most others measure the electrical properties of the soil (usually a property called the dielectric constant).  The electrical properties are strongly related to volumetric moisture content.

One soil moisture measurement system that has been widely used in the past is the tensiometer.  This is a tension measuring system consisting of a 300 mm (or 600 mm or 900 mm) long probe with a pressure gauge.  Probes (with the gauge) cost approximately $250 each.  Although they have the advantage of low cost, they can be damaged by frost so you need to bring them in each winter and re-install them in spring.  Readings from tensiometers can be quite variable, and it is difficult to relate the readings to volumetric moisture content.

While tensiometers are still a workable system, it is worth considering the range of newer soil moisture systems available including:


A dielectric constant measuring system consisting of a three-metre long ‘strap’ that can be installed horizontally or diagonally in the soil.  Probes cost approximately $795 each, and the reader costs around $560.  Advantages include the large measurement volume and accuracy (and it also provides a soil temperature measurement).  The disadvantage is cost.  The probe is not portable – it needs to be permanently installed at a site.

Campbell Scientific Hydrosense (Model 620)

A dielectric constant measuring system consisting of a pair of 120 mm or 200 mm long metal probes connected to the electrics box at the top, and a reader.  The system costs about $1,450, including both probes and reader.  The 620 model is portable, as the probes can be moved from site to site.  See for more information.


A dielectric constant measuring system consisting of a 200 mm or 100 mm long probe shaped like a ruler.  Probes cost $400 each if you are just buying a few, but drops to $160 for a larger order.  Handheld readers are available at $950, or data logging systems for downloading to computers from $840.  See for more information.

The New Zealand Irrigation Manual describes other options, and you will be able to find more information on available systems from suppliers of soil moisture instruments.

It is becoming more common to use data logging systems (rather than hand-held readers) with soil moisture measurement probes.  They will take a soil moisture measurement automatically once a day (or whatever frequency you choose) and store the result.  It is then easy to graph the daily readings over a time period, to watch the soil moisture respond to an irrigation or rainfall event, and to identify seasonal trends and patterns.  Of course you can also manually graph daily readings if you have a hand-held reader.

For some technologies, the expensive part is the power supply and reader.  If you have some skills in electronics and circuits, there are a few ‘do-it-yourself’ style options that can significantly cut the cost.  For example, the CS620 can be read using a multimeter which measures frequency.  The EcH20 can be read using a basic $40 multimeter, and you will also require a regulated power supply (such as an old car battery and a $30 DC/DC regulated transformer).  With those options, you won’t be able to use the calibration curves that are supplied with a full soil moisture measurement system (see next section) so you will need to put in a bit of extra work to interpret the readings from your multimeter.

But what do the numbers mean?

The various probes respond to soil moisture in different ways, and, fundamentally, the reading is just a number that is related to the dielectric constant (or other property) of the soil.  However you don’t need to worry about this if you buy a full soil moisture measurement system, complete with reader, as it will come with calibration curves.  These convert the readings from your system, and your soil, into the common language of volumetric moisture content.  Once you have soil moisture readings in these units, it is easy to set the soil moisture threshold at which you will turn on your irrigation.  You can compare your soil moisture readings with other growers as long as they are also using systems calibrated to volumetric moisture content.  However it is not meaningful to compare ‘numbers’ with a neighbour who is using tensiometers or some other uncalibrated system.

Either way, it is important for you to come to an understanding of what the numbers reveal about the water needs of your trees.  When you first start water monitoring on your property, establish two critical soil moisture levels, as measured by your system.  You’ll need to know the field capacity and the wilting point.  As a plant uses up water, the soil drops from field capacity down to wilting point.  Initially, the water is weakly bound to the soil particles, and is easy for the plant to obtain; this is called ‘readily available’ water.  As further water is removed, it becomes more difficult for the plant to access what remains – it’s called survival water.  We aim to keep soil moisture within the range of readily available moisture.  Usually the soil will ‘fill’ over winter.  When measuring soil moisture at the end of winter, wait for a couple of days after rain so that the free water has drained out and the reading should correspond to field capacity for your soil.  It is worth measuring on several occasions to check that you have it right.

To measure wilting point experimentally you need to be a bit hard-hearted. Turn off one of your micro-sprinklers during summer so that a tree misses its irrigation until a few leaves begin to turn yellow – that’s the wilting point.  However, there is a less sacrificial way of doing this – a soil moisture expert is able to estimate wilting point for you from a knowledge of your soil type and its texture.  If you have purchased a modern, calibrated soil moisture system, your supplier will help you with the establishment of these critical moisture levels, and with understanding how the readings are related to the tree’s water needs.

7.4 Irrigation scheduling

If your system is calibrated to volumetric moisture content, it is easy to establish a first estimate of the ‘refill’ point (at which you will turn on your irrigation).  Generally for walnut trees, it is accepted that the readily available water range is 50 – 60% of the total range between field capacity and wilting point.  So if you know the volumetric moisture content at field capacity (call this F) and at wilting point (call this W), then you can calculate refill point (R) as follows:

R = F – (0.5 x (F – W))

Let’s take an example.  If the volumetric moisture content of your soil at field capacity is 30%, and at wilting point is 10%, then:

R = 30 – (0.5 x (30 – 10))
= 30 – 10
= 20%

(N.B.  Do not try to do this calculation if your readings are not calibrated to volumetric moisture content.)

When your soil moisture reading drops to 20% (volumetric) in the above example, you begin to irrigate.  As you gain experience, you will probably refine your initial estimate of the refill point (R).  As you observe your trees responding to various soil moisture levels, you will be able to see for yourself at what point it becomes difficult for them to extract water from the soil.

The purpose of irrigation is to bring soil up towards field capacity.  On a heavier soil, don’t bring the moisture content all the way back to field capacity – you should leave some space in the soil in case it rains.  If you are on a very light soil, however, it can be risky to leave such a buffer, because the soil holds so little water, even at field capacity.  If you have a pump breakdown, you may only have a few days before your trees become water stressed.

If your system does not come with calibration curves, you have two options.  One is to calibrate it yourself[4], and the other is to leave it uncalibrated but determine a suitable refill point by experience.  The uncalibrated approach is still a very useful option – your soil moisture measurement system will ensure that you stick to a consistent and objectively-measured refill point.  You begin by establishing field capacity and wilting point by taking readings with your probes at the end of winter and at ‘leaf-yellowing’ point.  (Note that tensiometers cannot read down to the very dry conditions at wilting point).  Then the simple way of scheduling your irrigation is to never allow soil moisture to get higher than field capacity, and never as low as wilting point.  You should establish a refill point  (for your measurement probes) at which you will turn on your irrigation.  Your refill point should be a comfortable margin above wilting point, and well below field capacity.  You will be able to refine the initial estimate with experience.

When you first install your soil moisture measurement system, it is a good idea to take readings daily, and record rainfall.  You can then plot your soil measurements on a graph, draw on the rainfall events, and check that the soil moisture is responding in a reasonable way to rainfall events, and to periods with no rain.  Once you are up and running with your system, you may only need to take a reading once or twice a week.

[1] Most newspapers report the E.T. figures daily or weekly.  They are measured for a grass sward, but are a useful indicator for your orchard.

[2] When soil is at field capacity, it is holding the maximum amount of water possible in its micro-pores, but the water has drained out of its large macro-pores, leaving them filled with air.  Thus there is water available for plant growth, but the soil is not saturated.

[3] Volumetric moisture content – the percentage of each cubic metre occupied by water.

[4] Calibrating soil moisture probes yourself is not straightforward, but is possible.  You would use your probes to measure soil moisture content at a range of values.  Take samples of the soil (samples of known volume) at all these points, which you weigh, then dry, then weigh again, to establish the volume of water that was in the known volume of soil.  This enables you to convert between the readings from your probes and volumetric moisture content.


This section does not appear in the printed version of the manual. It has been added to reduce the amount of screen space required for the table of contents. The text is exactly the same, but four of the chapters have been renumbered and are now in this new section. 

CHAPTER 1 – A seasonal programme

1.1 Winter

Training for all but very small new trees

Do not leave it too late because walnuts are ‘bleeders’ when the sap starts to flow.  Choose a dry day for training or pruning.  In the climate of Central Otago, growers recommend pruning all walnut trees in summer because the heavier frosts in winter and more extreme temperature swings cause winter bleeding to be more severe.


Winter is a good time for replacing any trees that have died, both shelter trees (deciduous/frost tolerant) and walnuts.  Determine, if you can, the reason the original tree died.  If it died from Phytophthora root rot, there is no point in putting another one in its place until you have fixed the problem.  You may need to improve the drainage by ripping, or sterilise the soil, or dig it out completely and replace it with non-infected soil.


If you like year-round bare ground under your walnut trees, you may want to spot spray some of those recalcitrant perennial weeds before the early spring application.

Cleaning up and mulching

Clean up the prunings.  If you are spraying with a pre-emergent spray like simazine, clear the drip zone of trash (like last season’s husks, branches, and dead weeds), since the simazine needs to adhere to soil particles in order to work.  (If you are considering organic certification, bear in mind that using simazine may extend the period you have to wait.)  Prunings can be mulched with a slasher mower on the back of a tractor if they are thrown into the service areas between rows.  Mulch can go onto the drip zone area in late winter too.

Winter fertiliser

Apply the winter fertiliser but not too early if you typically have a wet winter because some of the more soluble fertilisers, such as nitrogen and potassium, may be leached out and lost.  On the other hand, do not leave it too late, because the roots will probably begin growing before the leaves emerge.  There’s usually plenty of water in the soil in the spring, but if not, and you have irrigation available, it may pay to water the trees – but be careful not to overwater, as you could attract problems with Phytophthora.

Late winter/early spring micro-nutrient application

Check if micro-elements are necessary, usually working on the basis of your late summer leaf analysis, and/or by consulting a horticultural advisor.  If you do need to add micro-nutrients, then you will also need to decide whether to spray them on as a foliar application (e.g. with your first copper spray), or spread them on the ground.

1.2 Spring

Spring weed spray

It’s a job for early spring.  Apply herbicide spray to the drip zone – this may be either a pre-emergent spray (like simazine) applied to bare soil, or a foliar spray (such as glyphosate) applied to spring growth.

Shelter tree planting for some species

Another job for early spring.  This is a good time to plant native shelter species, or species such as tagasaste that are somewhat frost tender when small.

Copper spray against blight

A job for mid-spring. There is a great variation in time of bud-burst between varieties and also between districts and seasons.  The most important decision is when to start spraying copper (with a penetrant) against walnut blight.  The best current advice is to spray when around 80% of the buds have started to burst – you’ll see the bracts begin to crack open and pull back from the bud).  Keep a close eye on this process, particularly around the second week of September, but it can vary by a few weeks, either side.

The first green tissue to emerge is the male catkin, and it is well worth covering it with copper.  The other reason for spraying early is that copper can be phytotoxic (burn the plant in strong sunlight) on tender young tissue such as leaves or, more worryingly, on the receptive part of female flowers.  By applying the spray early you can safely use a high rate of copper and apply plenty to run into the cracks and crevices.

Protect against frost

Cover very small trees (particularly if they are still in their shelter boxes) with frost cloth or other coverings during October and November (if your region is likely to have frosts in these months).  If the trees do get frosted, don’t panic; most will sprout back from later, or secondary buds.  It is a good idea to spray frosted trees with copper to discourage blight entering wounds.  The best orchard floor to reduce frost is firm and flat.  Bare ground absorbs heat during the day and lets it out at night so at the very least, keep the grass cropped short at this time of the year.

1.3 Summer


Summer is a relatively quiet time of the year.  You may need a second weed control spray if using foliar sprays rather than pre-emergents.  This is particularly critical if your trees are mature, and you expect to harvest a crop.  Most growers will mow the service area at least once, and, for some, many times, depending on their desire for tidiness.  Copper sprays should be kept up until at least January, though the frequency depends on the weather (every two weeks or so if the weather is warm and damp, but you can hold off if it is dry and/or cold).


Irrigate as required, working from your soil moisture measurement equipment, or the advice of your soil moisture monitoring service provider, and taking rainfall into account.  It is worth getting this right – there are penalties for both under-watering and over-watering.

Tree training

Trees in their first year may be trained in late November or so – after the greatest frost danger is over.  Growers in Central Otago may prune all their trees at this time.

Leaf analysis

Collect your leaves for a nutrient analysis in the last week of January or early February.  It’s the time the laboratories have used for their standard results and they’ll be comparing them with the results of your samples.

Preparation for harvest

Prepare the drip zone for the nuts to fall on.  A tidy, flat surface will make an enormous difference when you begin to pick up the nuts.  For the safety of our consumers, don’t spray weeds close to harvest.

1.4 Autumn


Pick up the nuts every two to four days, but every day if it’s wet.  If the nuts are left too long on the ground, or you don’t keep them dry, mould will develop and the processor will reject the kernels.  In large orchards in USA and France, tree shakers are common, and it is likely that larger orchards in New Zealand will use the same method.  A tree shaker allows you to shake the tree just once, or maybe twice, during the season, and pick nuts up just at those times.  You would also need a machine to remove the husks.

Late spray

A job for late autumn/early winter.  If you have chosen to maintain bare earth over the winter, after harvest is a good time to spray perennial weeds, as they ‘close down’, using a foliar spray.

Flowering, pollinating and fruiting

See the coloured pages ‘The reproductive life of the walnut’ for photographs of pistillate flowers, catkins and fruit development.  You will find descriptions of each popular cultivar in Section 2, Chapter 3, including their specific flowering times and other characteristics.

2.1 Flowering and pollination

Flowering is the first step in the reproduction of flowering plants, which in walnuts produces the nut.  Successful nut production in walnuts requires pollination, that is the fertilisation of an embryo with pollen (the male zygote).

Flowering cannot take place while a plant is still in its juvenile stage of development.  A young seedling walnut tree goes through a transition phase in which flowering begins to occur, but it is not until tree grows into its mature phase that flowering is extensive.  The length of time the juvenile phase lasts varies between cultivars.  Grafted trees have the advantage of reducing the time taken to come into fruit, because the scion wood is already several years old at grafting time. Commercial growing of walnuts from seed is not recommended because of the extra number of years before harvest and also because the seedlings will lack uniformity of productivity and quality.

In walnuts, the male pollen-bearing (staminate) flowers and female nut-producing (pistillate) flowers are borne separately on the same tree.  For most cultivars pollen production occurs before (protandrous), or in a few cases after, pistillate flowering (protogynous). Walnuts are self-fertile although, within a single cultivar, the period of overlap between pollen production and pistillate flowering is usually quite short, so we recommend that you include pollinator cultivars in your orchard along with your main crop cultivars. The most popular selected cultivars, Rex and Meyric, are able to pollinate each other, but many experienced growers prefer to plant about five per cent of other cultivars to be sure.

Pollen shed times for different cultivars vary between about mid-October and mid-November, depending on local climate.  Walnuts are solely wind pollinated, and pollen is able to travel long distances, remaining viable for up to two days after being shed. Even so, plant your pollinators in the same block as your main cultivars.

Male flowers

Male (staminate) flowers are grouped in hanging clusters called catkins.  A single catkin contains as many as 100 to 160 individual pollen shedding male flowers.  The catkins form in one year and shed their pollen during the next.  They form in the axils of the leaves, immediately after budding.  At that stage they are less than one centimetre long, greyish, and cone-shaped.  They gradually change to green over the summer.  During autumn they change to brown and stop growing during winter. 

In spring, the catkins start to swell and lengthen.  During this stage the pollen grains are formed.  The colour changes to light green and you will be able to distinguish individual male flowers.  Over a couple of weeks the catkin loses its rigidity and begins to hang downwards. The individual male flowers then begin to separate.  The anthers, from which the pollen is shed, turn yellow.  Beginning at the base of the catkin, the flowers open and discharge their pollen.  A single catkin discharges all of its pollen over the space of a few days, depending on climatic conditions.  The catkins then turn brown, dry up and drop off the tree.  Higher temperatures cause more rapid catkin development and pollen shedding. 

From the time the catkin begins to hang downward and open, it is susceptible to frost damage and being blown off the tree by high winds.

Female flowers

Female (pistillate) flowers originate in a bud situated at the tip of a shoot.  In most cultivars they are terminal buds, but some cultivars also fruit on the end of spurs growing laterally from the main branches.  When a season’s shoot growth is complete, pistillate flowers begin to form inside some of the buds.  The following year they emerge in spring just after the appearance of new leaves.  They are usually grouped in clusters of two or three.  The female flowers are small, round and green, with feathery stigmas emerging from the top.  The stigmas capture the wind-borne pollen and are able to accept walnut pollen while rejecting the pollen of other species.

When they first appear, the flowers are not receptive to pollen.  Then the stigmas gradually change from red through to yellowish green.  The stage of optimum receptivity is when the stigmas are well developed and are yellowish in colour.  They have a moist, glistening appearance caused by a secretion on the surface.  While the secretion is there the pollen is retained on the surface and, if viable, will germinate shortly after contact and a resultant pollen tube will grow down to reach and fertilise the embryos.  If the flower is not pollinated during that period, fruit will not set.  A single flower normally remains receptive for four to six days.

After the receptive period, the stigmas bend completely backwards and become paler. Next the stigmas develop brown streaks and begin to dry up and blacken.  The flowers are then termed post-receptive.

2.2 Fruit development

For the first two to three weeks after flowering, both pollinated and un-pollinated pistillate flowers grow at the same rate.  But after this period, it is only the successfully fertilised nutlets that continue to grow, while the others fall off.  The fruit grows quickly for the first seven to eight weeks.  It then grows more slowly for another two to three weeks to achieve its final size in approximately mid January.  At this stage the developing kernel is watery, and the shell inside the husk is either soft or hardening depending on the cultivar.  The shell continues to harden over the next few weeks, and the kernel continues to develop.  The nuts are ready for harvest about 20 weeks after fertilisation, in about early April, depending on cultivar and region.

CHAPTER 3 – Money matters

How much money can you make growing walnuts?

It’s not an easy question to answer, because we don’t yet have enough growers producing commercial quantities. But we can give you some indications.

Let’s begin with a simple formula:

The sections below provide more detail on the Income and Expenditure parts of the formula. But first, there is one more consideration – lead time.

3.1 Lead time

Walnuts are relatively slow to start cropping.  You may have your first small harvest when your trees are around 6 years old (and remember that you may have spent several years establishing shelter before planting your walnut trees).  It’s unlikely that you will have significant income until your trees are 10 years old.

The graph on the next page shows the average yield per tree at the Lincoln University trial block, for years 5 to 13 after planting.  The averages have been taken across the 11 highest yielding cultivars in the trial (including Rex and Meyric).

Your orchard will reach full production once the tree canopies have grown to cover your full land area.  At that stage you must prune or remove trees to allow the sun to penetrate and help you keep up your production.

The length of time it takes for your trees to reach full production depends on many things – your local climate, the cultivar you choose, the spacings at which you plant your trees, and your management of them.  Lead time until full production is an extremely important aspect of the economic viability of your enterprise.[1]

Tree spacing has a strong influence on yield in the early years, though once the canopy is closed, it doesn’t really matter how many trunks are underneath, unless you let them get overcrowded so the light doesn’t penetrate.

The following graph demonstrates the effect of tree spacing on the early years. The black bars show what the yield would be if the trees were planted at 7´7m spacings. That’s 196 trees per hectare (ignoring headlands and shelterbelts).  In comparison, the white bars show what the yield would be if the trees were at 10´10m spacings (100 trees per hectare).  The yield per tree from the first graph has been multiplied by the number of trees per hectare to give production per hectare, for these two examples of possible spacings.


[1] However, in the early years, you will be operating at a loss, so you will be entitled to offset the loss against other sources of income on your tax return.

Clearly you achieve higher yields (in t/ha) in the early years with more closely planted trees, and your orchard will reach full production (full canopy) more quickly.  However, you must balance the higher production against the cost of the extra trees, and the cost and inconvenience of thinning them out later to allow the light to penetrate.

3.2 Income

There are many variables affecting income, so to give an idea of the possible range, we will give both a conservative estimate and an optimistic estimate.

At the conservative end, your yield each year may be two tonnes per hectare (in shell) once your trees reach full production. At the optimistic end, it may be nearer 3.5 tonnes per hectare each year. We will assume that the average crackout rate (the kernel as a percentage of the total weight of the walnut in shell) is 45%. The conservative production is 900 kg of kernels per hectare, and optimistic production is about 1600 kg of kernels per hectare.

The current[1] price paid to growers in New Zealand’s largest processing factory is $10.50 per kg of kernels for top grade and $6.50 for baking or second grade. We will use these as our conservative and optimistic prices. 

However, it’s also worth noting some other points on price. The international commodity price for walnuts means that U.S. growers are paid around US$0.6 per pound (say NZ$2.10 per kg) in-shell. We do not want to put New Zealand walnuts on the international commodity market. We need to compete on high quality and niche products, rather than on quantity, but it is worth being aware of the figure. At the upper end, some growers who crack their walnuts by hand sell them for about $20 per kilogramme of kernel. Note though, that you must factor in your labour, equipment and building costs if you plan to do your own processing.

If you are selling to a processor, your price will depend on your care in harvesting. If you can present clean, well dried nuts, the standardised grading system will ensure that you get the top grade and higher price. (See Chapter 11 Harvesting and Product Quality for more.)

The second grade price currently paid in Canterbury and the lower estimate of production give us a conservative view of your projected income:





$5,850 per hectare

per kg of kernels


kg per hectare


The higher production figure and the price for first grade kernels give a more optimistic view of your income:





$16,800 per hectare

per kg of kernels


kg per hectare


So the range for income is $5,850 to $16,800 per hectare at full production. (The mid-point is about $11,300.)

3.3 Expenditure

Your expenditure includes the costs establishing your orchard and the cost of operation each year. As a rough guide assume that half the cost of setting up is in the land, quarter of the cost in the trees and the final quarter in irrigation and other equipment.

Establishment costs (paid once at the start)

Let’s assume that you have bought your land and that you have electricity to run your pump. We can give you an indication of the other establishment costs here. However, treat this table as a shopping list, not a price list, because costs vary widely depending on many factors (for example, to what extent you employ contractors, or whether you do everything yourself).

Your establishment costs may include…




Implement shed and tractor shelter



Pump shed












Sprayer, e.g.,


    Knapsack sprayer



    Battery-driven sprayer






Cultivation (assuming you need a contractor)


per hectare

Ripping of shelterlines and crop lines


    Contractor using 4´4 tractor 80 to 100kW


per day

    Hire a winged ripper, rip to 60cm


per day

Grass seed


per hectare

Weed control


    Spray purchase


per spot

    Contractor with tractor, spray extra


per hour




Purchase shelter trees, e.g.,


    Alders, oaks



    Eucalyptus, wattles (in root trainers)



    Tree lucerne, willows (in root trainers)






Planting of shelterbelts


per tree

Purchase walnut trees



    Up to 25 trees


per tree

    More than 25 trees


per tree

    Selected seedlings for timber and nuts


per tree

Tree protectors, e.g.,


    Green plastic spray guards



    Corrugated plastic boxes



    Nova Flow lengths






Well and pipe liner


per metre

Pump, e.g.,


    Submersible pump (3 to 5 kW)



    Surface pump



Headworks (incl. filter, stop valve, flow meter, etc.)



Hire trenching machine


per day

Drippers for shelter trees


per tree

Alkathene pipe for laterals (13 or 16mm)


per metre

Individual sprinklers for walnut trees


per tree

LDPE pipe for submains (32mm)


per metre

Operating costs (costs each year)

Your operating costs are the costs of maintaining your orchard and continuing the development process each year.  We’ll assume that you will do the pruning of your commercial trees yourself, though it would be an additional cost if you hired a contractor to train and prune your trees.

Orchard expenses



Sprays for weed and blight control



Leaf or soil tests






Electricity for irrigation



Irrigation maintenance





per hectare

Tractor and mower maintenance


per year

Contracting (allowance for occasional spraying etc)


per year

Shelterbelt pruning (mechanical @ $175 per hour)


per year



per kilo




Office costs





Insurance for machinery, office, public liability



Rates (proportion)


Printing and stationery



Telephone (proportion) and tolls



Subscription to Walnut Industry Group






Interest on mortgage (@ 8.5%)


per $1000

Bank fees



CHAPTER 4 – Harvesting and production quality

You may have your first significant harvest after seven seasons. It will depend on your growing conditions and management including your tree training methods in the previous years.

It’s likely that your production will increase significantly each year for several years, until the tree canopies fill the available land area.  At this stage you will need to manage the tree canopies to allow the light to continue penetrating, otherwise your production will fall.  Management involves thinning out branches and possibly removing whole trees.

4.1 Preparation

Your walnuts are likely to be ready to harvest in early to mid April depending on your climate and cultivar. You will notice the thick green husks begin to split and the walnuts will drop to the ground.

Ideally the walnuts should fall clean from the husks – it’s been one of the criteria for the New Zealand selections. Even so, in very hot seasons, especially if temperatures are high between March and April, the husks can dry and stick on to the shells. It happened in Canterbury in 2001 and happens regularly during the harvests in Central Otago.

Before harvesting you should prepare the orchard floor and make some key decisions.

It’s important to ensure that the orchard floor is flat and free of stones and sticks that would be caught up with the crop when you harvest. You may need to mow the grass several times before you begin and using a roller a few days before the first walnuts fall will save you hassles and time later. Usually the first few walnuts to fall will be the diseased and undeveloped ones.  Go through with the mower to clean up the debris.

Decide whether to allow your walnuts to drop gradually over the three week harvest season or spray with Ethril, a chemical that encourages the tree to release them all 10 to 14 days later. The walnuts should not remain on the ground for more than a couple of days before you pick them up, otherwise they can become mouldy and discoloured.  To decrease the number of times you need to go through the orchard picking them up, it is usual to shake the trees.  While they are small, you can do it by hand or with a rubber mallet, but as they grow you’ll need a mechanical shaker which grips the tree trunk (or individual branches).

4.2 Harvesting methods

In the early years, at least, you may choose to pick the crop from the ground by hand. It’s hard work and you will probably be able to collect no more than 200 kilograms of walnuts in their shells in a single day. Once your production rises, you may find that you need a team of casual workers.

Even if you choose to stay with hand picking, a mechanical shaker would help you to make the most efficient use of your labour. Use plastic leaf rakes to scrape the walnuts into convenient piles. (Once you are raking, you’ll be glad that you put the time into creating a flat surface free of stones and debris.)

There’s a range of mechanical harvesters available around the world. They are commonly used in the large walnut orchards of California and France where most growers own a mechanical shaker and a harvester which operates like a street sweeper.  Because the NZ industry is in its early days, most orchards are relatively small and consist of young trees. In future, contractors may own the large, expensive equipment so that growers can call them in for harvest.  However in the meantime, there’s a range of options available for that in-between time when your harvest is too great to do completely by hand, but too small to justify investing in large mechanical equipment.

The hand-pushed harvesters offer a less tiring alternative to hand picking. This one has a series of discs that catch the walnuts and flick them into a tray. Another version has a plastic wheel with fingers to do the same job. Hand-pushed harvesters are light and maneuverable and you can use them in short grass. They don’t work backwards and can ride over the nuts if the crop is too thick on the ground, so you would need to use a leaf rake to spread them ready for collection. Hand-pushed harvesters cost around $1,200.

In Australia a tractor-mounted harvester uses discs in the same way as the hand-pushed models, but has a sweep four metres wide.

This is Henrietta, a suction harvester and the result of Kiwi ingenuity. Henrietta is a one-off production. She makes harvesting dusty and noisy work, but the suction works well when the crop is raked into rows. Henrietta has a useful added feature: a system to remove dust from the crop. You could make your own suction harvester or import one from Italy.

Another option is a device that combines a brush and conveyor on the right of the tractor. The collection system operates from the tractor’s power take off and the walnuts are sent to a bin that you can tip to fill bags.

4.3 Ensuring a quality product

Nut Industries of New Zealand defines top grade in-shell walnuts as clean and intact.  They must not be mixed in with foreign matter such as sticks and leaves. Top grade kernels have a honey blonde or light tan skin with white flesh, and should be free of mould and disease. Uniformity is important too. The easiest way to achieve it is planting grafted cultivars rather than the much more variable seedlings, and keeping nuts from each cultivar separate at harvest.

Although it takes some care to achieve top grade prices, Canterbury’s major processor, Cracker of a Nut, reports that growers who have been supplying walnuts for several years are able to get a high proportion of their crops into the top grade.

For top quality, harvest your walnuts at least every second day (though if you are shaking the trees and/or using Ethril, you will not need to do it so often).  Walnuts will deteriorate quickly if they are left on the ground for more than six days.

A research team at Lincoln University found large differences in quality when they compared nuts harvested after three days on the ground with picking the nuts from the tree to avoid them coming in contact with the soil. 97% of those picked from the tree had acceptable kernel colour, while only 72% of those picked up after three days on the ground did.  The researchers were using a cultivar with a weak seal, and the differences are likely to be less striking for cultivars with stronger seals such as Rex and Meyric.

Wash the nuts down with a hose or tumble washer and place them on racks to dry, turning them from time to time. Make sure that you keep them out of the rain or evening dew, but somewhere with air circulation. Preferably, set up a drying room with a forced-air drying system or send them to a processor who can dry them for you.

Effective drying is critical for achieving a high quality product.  Drying should bring in-shell walnuts to a moisture content of 8% by weight.  Wetter than this, and you risk mould; much drier than this and you risk rancidity (but note that air-drying on racks in New Zealand is unlikely to over-dry them).  Growers with large orchards in California and France usually have their own forced-air dryers and deliver the nuts to a processor once they are dry.  In New Zealand it is usual to dry walnuts using an air temperature of 30ºC or less.  The Californian Walnut Production Manual says that air at 43ºC or above will cause rancidity.

Walnuts will go mouldy if they are hot and have no air circulating, so don’t put walnuts in plastic bags or buckets.  Don’t wash walnuts while they are inside an onion bag or hang them up while they are still wet. Use a drying rack. Bear in mind that the walnut shell is porous, so the shell might look dry on the outside at the start of the harvest, but the nut will still be wet inside.

If you are sending your walnuts to a processor to be force-air dried, wash them first, dry them on racks for two days, put them into onion bags and deliver them as soon as possible.

Forced air drying produces superior results. Researchers at Lincoln University cleaned walnuts by water blasting, then dried them in a seed dryer with the forced air temperature not exceeding 30ºC.  They compared this treatment with the traditional small-scale method of hanging walnuts from rafters in onion bags. The differences are startling.


Drying in onion bags

Percent meeting ‘good’ standard

Forced Air Drying

Percent meeting ‘good’ standard

Shell colour1495
Kernel colour5393
Kernel mould8398

You should be wary of rodents and other pests while your walnuts are laid out to dry or in storage.  Not only can they damage your crop but also leave droppings amongst it. (You can imagine the damage to our crop’s reputation.)

4.4 Nut Industries of New Zealand Quality Mark

Nut Industries of New Zealand has written quality standards for processors of walnuts and hazelnuts.  The standards are to ensure high quality nut products in New Zealand. Processors can apply to be accredited to the standard and, if successful, can then use the NutNZ Quality Mark on their product. (Contact NutNZ for more information.)

If you are supplying a processor who is accredited to the NutNZ Standard, your walnuts must be of sufficient quality for the processor to meet the standard.  The important characteristics are:

  • Freedom from foreign matter (sticks, leaves, other objects and particularly any kind of animal or bird droppings)
  • Intactness/freedom from damage
  • Freedom from mould, disease and rancidity
  • Cleanliness
  • Normal development
  • Uniformity


CHAPTER 1 – The research

1.1 Introduction

Our research projects are a vital part of NZWIG’s contribution to the success of the walnut industry.

In the past, the New Zealand Tree Crops Association, Lincoln University, Nut Industries of New Zealand (formerly Southern Nut Growers Association) and individual people and businesses have developed a good body of knowledge on how to propagate, grow, harvest, process and market walnuts. However, there are still many questions we can’t answer, and ways that we could improve.  NZWIG believes it is essential for us to be continually improving the quality of our product, our efficiency and profitability in producing it, and our ability to manage threats such as pests and diseases. 

This chapter will tell you about the research and development projects currently underway, and the latest results and observations emerging from them.  We will update it from time-to-time.

1.2 Managing walnut blight

Issues and aims

NZWIG’s flagship research project is on managing walnut blight. Blight is caused by the bacterium Xanthomonas campestris pv. juglandis and is the main disease challenge to walnut growers in New Zealand.  Both in New Zealand and overseas, the primary method of control for walnut blight is spraying with copper-based products.  The spray is a significant cost to growers and does not completely control the disease.  There is also a well-founded concern that, over time, copper could build up in the soil and decrease the diversity and resilience of soil ecosystems.  In addition, in both Europe and the United States, copper resistant strains of the disease are now appearing.

During the last three years we have found that Mankocide provides significantly better control than the previous standard, Kocide.  Mankocide is a combination of Kocide and Mancozeb, which is a dithiocarbamate type of bactericide/fungicide.  Mankocide costs about the same as Kocide, so it’s a significant advance; however, Mankocide is not the ‘ultimate’ solution since consumers are increasingly resisting such chemicals. 

Ten years ago, organophosphates were a significant part of the pest management arsenal of the NZ pipfruit and kiwifruit industries, now the industries only permit limited use of the most benign organophosphate.  The change has been driven by consumer demand for residue-free fruit and for quality assurance that specifies very sparing use of organophosphates.  Dithiocarbamates are now appearing on the consumer’s radar and are likely to be put under a similar spotlight.  Fortunately, our final sprays are typically applied four months before harvest and the edible part of the walnut (kernel) is protected by both a shell and a husk from the sprays[1]. Even so, we need to accept that, in the medium to long-term, we will need to use dithiocarbamates sparingly.  For these reasons, our research work aims to reduce the amount of spray chemicals required for blight control by careful timing of applications and by trialling alternative spray types.

We are investigating:

Optimum timing of bactericide sprays to maximise the effectiveness of the spray and minimise quantity of copper you will need to use. Working out the best timing involves considering the physiological stage of the walnut buds, leaves and reproductive parts, and also climatic conditions such as temperature and humidity that strongly influence the blight’s development.

Alternative spray types, including low-dose copper sprays, host-response activators (that stimulate the tree’s own defences) and so-called ‘soft’ bactericides whose environmental impact and risk to sprayer and consumer is minimal.  Being able to alternate spray types through the season could lead to reduced use of copper and reduced risk of copper-resistance in blight bacteria.  We are testing the performance of different sprays, and plan to apply for registration of any that prove useful.

Biological control with bacteriophages to make use of the natural “predators” of blight bacteria, by boosting their strength and speed of action. The project involves isolating bacteriophages (viruses that attack and kill the blight bacteria) from the canopies of walnut trees, understanding their lifecycle, and selecting the strongest.  The idea would then be to mix a “cocktail” of the most virulent, and spray it back onto the trees.  The bacteriophages are entirely specific to walnut blight bacteria, so have no effect on any other living thing.  Biological control is very much a future-focused research project, looking to the options available to walnut growers in 10 or more years.

[1] The permitted residue limit is currently 7 parts per million. The tests we performed to gain registration for Mankocide failed to detect any residue. (The detection limit was 0.2pm.)


The research on walnut blight is a collaborative effort with growers and researchers in Tasmania.  Spray timing and alternative spray experiments are conducted in commercial orchards both in Canterbury and in Tasmania.  Dr Tim Jenkins is leading the New Zealand input. 

Visually assessing nuts is an effective way of measuring the level of blight control.  We take account of the seasonal change in the number of affected nuts in the canopy. (Visual symptoms appear around the start of December and increase from there, but severely damaged nuts start to fall off in January.) We are continuing to experiment with bud sampling and bacteria population counts to give an ‘early warning’ of the disease before the obvious symptoms are present on nuts etc. The benefit of the bud population early warning is as a potential ‘spray trigger’ but the approach is much more expensive than visual analysis and not always as well correlated with the ultimate test – the loss of yield.

The focus over the next season will be on further determining the best timing of the sprays.  For example, how important is it to reduce bacterial numbers at the start of the season, i.e., is it worth following the (essential) bud-burst spray with a follow up spray (or even two)?  We are convinced that subsequent spring sprays are best tailored to the climate (blight grows strongly in warm, humid conditions), and will trial a variety of approaches.  We are able to use results from both our climate and the more humid Tasmanian climate.

We will also be trialling new chemicals and combinations of treatments.  Horticulture in general is demanding so-called ‘soft’ bactericides and fungicides which are more environmentally benign and don’t have residue implications for consumer. We hope to be able to trial some of that new technology.

The bacteriophage work is very specialised, and is being undertaken by scientists from HortResearch under the leadership of Dr Monika Walter, based at Lincoln.  Taking the fundamental knowledge gained by Dr Sandra Romero during her PhD studies, the team will be moving towards laboratory and then field trials of phage cocktails to manage blight.

Project duration and support

The project has received continuing funding from both the New Zealand government (through the MAF-administered Sustainable Farming Fund) and the Federal Australian Government (through Horticulture Australia Limited) providing a total budget of about $150,000 p.a. (covering both the NZ and Australian research work).  The budget includes a cash contribution of $5,000 p.a. from NZWIG.  Further, NZWIG members, and in particular the research committee, are committed to an in-kind input (we give our time) of $28,000 p.a.  The first three-year period of funding ran from 2001 to 2004, and our new funding will give us another three years – to 2007.

Summary of results and recommendations from the latest research

So far we have completed three seasons of experimental work and produced these observations.

Mankocide provides better control than Kocide and Kocide is better than all the other alternatives and better than nothing.  The experimental results have clearly shown that Mankocide is most effective both on a total control perspective and control per kg of copper applied basis.  Mankocide is a combination of Kocide and Mancozeb.  We have tried Mancozeb alone but it does not give good control.  We suspect that the copper and dithiocarbamate act together to provide the improvement through a ‘multi-pronged’ attack on the pathogen. 

Timing of the first copper spray in spring is critical.  Spray at bud burst.  The blight bacteria over-winter primarily inside the buds and catkins, and it is from there that they invade new growing stems, leaves, pistillate flowers and pollen in the spring.  It is therefore important to kill off as many of the over wintering bacteria as possible before warm, damp spring conditions allow a population explosion.  However, copper spray cannot penetrate closed buds, so wait until a good proportion of the buds begin to open. See From Bud to Leaf in the coloured pages for the stage at which the copper can penetrate the bud in spring – just as the outside bracts begin to pull back.  This stage occurs around mid to late September in Canterbury.  It is worth keeping a close eye on your trees from the beginning of September onward, to ensure you get the timing right for the first spray. 

A second (and even third) early season spray may be a good policy.  Current thinking is that the early stages of growth, from bud burst through until when leaflets start to form on (say) 100mm long shoots, provide a good opportunity to reduce bacteria numbers to very low levels and reduce the disease pressure later in the spring.  If the bud-burst spray has been highly effective then spraying at this time is not necessary, but since buds burst at different times, the follow-up spray may be beneficial. It is much easier to achieve full coverage of new season’s growth early in spring than once the canopy is fully formed.

High temperature and humidity are the danger factors for blight growth.  Moisture on plant parts mobilises the bacteria, allowing them to invade new tissues.  Dampness also protects blight bacteria from desiccation.  The optimum temperature for blight population growth is 28 to 32°C, but it will grow, and black lesions will form, at temperatures above 12°C.  The weather in a Canterbury spring typically varies between warm-dry nor’westers and cold wet southerlies with some cool, dry nor’east sea-breezes in-between when anti-cyclones dominate.  All of these weather patterns are generally either too dry or too cold for highly accelerated blight development.  However, we do get periods of warm humid weather (and hence highly accelerated blight growth) usually brought on by a ‘low’ to the east of the North Island bringing warm, wet tropical air to Canterbury (and not over the Alps and dried out as in a nor’wester).  We suggest you look out for such weather patterns and time your sprays to coincide with these patterns.  We will be experimenting further with spraying before and after such events.

Blight bacteria can enter easily into wounds caused by frost, pruning cuts or broken limbs. It is advisable to spray immediately after an event like this unless you are quite sure the weather conditions are too cool or dry for blight development.

It’s safe to stop spraying in late January.  By then the shells are hardened and blight won’t penetrate the shells – though you may notice some black lesions on the husks.

An autumn “clean-up” spray is probably not helpful.  Spraying just prior to leaf fall, or even during the winter, can indeed reduce blight numbers on the outside of the buds.  However, the spray cannot penetrate the tightly-closed buds that will shoot next year.  It is better to get your bud-burst spray right – to kill bacteria on the outside of the buds and penetrate buds that are beginning to open.

Blight is unlikely to survive in dead, dried-off prunings if they are left on the orchard floor. Even if a few bacteria did survive there is almost no way for them to get back into the tree canopies.  Bacteria are mobile in water (e.g., dripping down through a tree canopy in rain drops) and on flying pollen grains, but they do not survive exposure to dehydration or UV light.

The cultivar Rex has consistently lower blight numbers than Meyric (and lower nut loss). We noticed a significant difference throughout several years and in many Canterbury orchards, and expect this would also be the case elsewhere in NZ, at least in similar eastern (dry) climates.  Meyric, in turn, seems less susceptible than other cultivars such as Dublin’s Glory, Vina and Tehama.

Our researchers have isolated several virulent bacteriophages – viruses that attack and kill walnut blight.  Each bacteriophage is capable of killing several strains of walnut blight bacteria in laboratory conditions, but has no effect on any other bacteria or other organisms.  Bacteriophages occur both in the canopy and the soil, but only canopy bacteriophages appear able to survive the dry and high UV conditions in the canopy. The most effective canopy bacteriophages are under further investigation.  Means of propagating, storing and deploying them in the canopy are being developed but there are still many challenges to overcome.

1.3 Other research projects

NZWIG is working on five research projects in addition to the blight project.

New cultivars

In New Zealand we currently have two extremely good walnut cultivars selected for use in commercial orchards – Rex and Meyric.  They have low susceptibility to blight compared to many other varieties, and they bear well.  They have excellent quality characteristics for processing.  In particular, their crack-out percentages (weight of kernel as a percentage of the total weight of the nut in-shell) are very good – so they are very efficient for both harvesting and processing.  The shells are strong enough to stay intact during harvesting or transport, under a weight of other nuts, but not so strong as to make cracking in the factory difficult.  They have good flavour and storage characteristics.

However NZWIG believes that it would be beneficial for growers to have a larger set of excellent cultivars to choose from. This gives us greater resilience to climate fluctuations, pests and diseases, rather than being totally dependent on a small gene pool.  In addition, there is a characteristic known as lateral bearing, which is valuable in that it provides earlier cropping when the trees are young or when they have been pruned back.  Rex is lateral bearing, but Meyric is not.  We aim to find lateral bearing cultivars that are at least as good as Rex and Meyric in processing quality.

It is no small exercise to select new cultivars that are excellent in both growing and processing characteristics, since it requires extensive screening of nuts, and then grafting, growing and monitoring the candidate cultivars for 10 to 15 years.

We have selected nine candidate trees for trialling.  Scion wood has been taken from these trees and has been grafted by Darrell Johnston of River Terrace Nurseries, Brightwater.  Darrell will grow the grafted trees on until winter 2005, when we will plant them at our trial sites on commercial orchards.  There are four trial sites, one near Wanaka, two in Canterbury and one near Wanganui.  The four sites provide a good range of climates.  At each site all candidates will be represented alongside Rex and Meyric. We expect to need 12 to 15 years to fully assess the candidates, but we should be able to start drawing conclusions in 8 to 10 years.


We are beginning to investigate the use of Paradox rootstock (Juglans hindsii pollinated by Juglans regia) which is likely to be more vigourous and help growers on poorer soils produce a commercial crop sooner.  Paradox rootstock is commonly used in California and we use Juglans regia in New Zealand.  Bernard Vavasour from Awatere Valley in Marlborough imported some Paradox-rooted Serr from California in the eighties and they out-performed regia-rooted Serr of similar vintage.

Darrell Johnston is growing Paradox rootstock from nuts and seedlings sourced from Bernard Vavasour.  The rootstock will start to become available for trialling from 2005 onwards and will be grafted with Rex and compared alongside conventional (regia-rooted) Rex at several trial sites.

Ross Jamieson at Quality Tree Company has also grown some Paradox rootstock and grafted Rex onto them. They are being trialled alongside regia-rooted Rex in a commercial orchard.  Ross has planted more hindsii nuts from trees surrounded by regia which have a good chance of producing a high proportion of hybrids.  The hybrid seedlings will also be grafted up and trialled in subsequent seasons.

Tree training/pruning and elevated Boron

We are comparing three different training regimes on Rex across three sites in Canterbury.  The regimes include the conventional technique (described in this manual), a very minimal training regime and a mid-way regime (also described in this manual).  In addition, we are testing whether elevating Boron up towards 200ppm will enhance apical dominance. Current thinking is that Boron should certainly exceed 30ppm (in foliage analysis) but there may be benefits of elevating the level further.  These treatments were initially applied in winter of 2004 and we will repeat them in subsequent winters.  We will monitor them by taking measurements immediately before winter training/pruning.

Management of the canopy in a mature orchard

We are using the Lincoln University trial block to investigate how best to manage the canopy in a mature orchard.  The oldest trees there are now 20 years old and were planted at 6m centres so the canopy is well closed and we believe many trees receive insufficient light for maximum production.  We are experimenting by removing of a couple of trees and significant numbers of limbs on other trees to provide more light to parts of the block.


Benchmarking is a practice used in a diverse range of industries. At its most basic, benchmarking is designed to help us learn from others by comparing our management practices and orchard performance (growth and yield) with other orchards. Horticultural examples of successful benchmarking include highly specific projects such as predicting apple yield from spring temperatures, through to more general projects on Australasian grape production and on the financial performance of vineyards.  This work is being carried out by Robyn Adams and Jane McKenzie (with guidance from Dr Tim Jenkins) under a one year Royal Society of New Zealand Teacher Fellowship.

They are measuring tree growth on a variety of orchards, as well as management factors influencing tree growth. Beginning in Canterbury, the walnut orchard benchmarking project is comparing block with block and orchard with orchard.

By comparing the wide range of management practices we can produce results very quickly, rather than collecting data from an experiment that would take more than a decade if we started from scratch. Even in the first year of this project we will have begun to define best practice for some of the establishment and management techniques for our trees.  NZWIG will facilitate continuing benchmarking and measurement on commercial orchards after Robyn and Jane have completed their project, working from the base they have established.

Our walnut orchard benchmarking project will provide:

  • Best practice models for establishing and maintaining orchards. It will provide a framework for you to compare the growth and productivity of your trees with those of other orchards with similar characteristics.
  • Specific measurements you can use to assess your orchard’s performance
  • Ways of assessing management techniques so that any orchard can contribute to future data and compare itself with best practice models – and so that we can improve the best practice models.
  • Information about a range of issues including factors that influence walnut blight.
  • The ability to convert good luck, bad luck and doubtful decisions into decision-making tools.

Data collection is underway at the time of writing, and you can expect results and recommendations to appear in the next update of Hot off the Press, as well as in upcoming newsletters and field days.

1.4 Disseminating information

To make an impact on industry profitability and sustainability, research and development results must reach the growers.  NZWIG is determined to make available current knowledge on best practice by means such as the NZWIG Information Pack, the Walnut Grower’s Manual, field days and our website ( We will disseminate new recommendations from research with technical articles in Health in a Shell, and by updating this chapter.

1.5 Supporting “industry good” R&D

We need to make research and development for the common good a continued focus.  There are two forms of support.  One is our own in-kind contribution of creativity, labour, and orchard space and the other financial support that allows us to buy science expertise or materials.

There are two sources of financial support:

  • Government funding – which we can apply for and use to purchase research services and materials in our own projects, and also funding of research programmes in other organisations that have flow-on benefits to us.
  • Industry levies – such as our existing, but voluntary, propagators’ levy system – it’s a small addition to the price of each tree. The money goes into research and is currently contributing to the walnut blight project.The dairy industry is a prime example of a large and successful industry-good research programme based on a commodity levy. It is charged per kilogram of milk solids produced. 

Accessing government research funds

The government recognises the value of research into agricultural production – perhaps not as much as some would like, but nonetheless, research funding is available.  In general, government support is only forthcoming when the user industry demonstrates the need for the research by contributing to the cost.  Industry contribution is usually made up of in-kind and cash contributions.

Government funds are channelled through various programmes such as the Sustainable Farming Fund, Technology New Zealand and the Foundation for Research Science and Technology’s (FRST) Public Good Science and Technology scheme.  In some cases the application for funds is expected to come from the industry with support from research service providers (universities, Crown Research Institutes and others), and in some cases from research providers with support form the industry.  Either way, the application will be assessed for its fit to government research priorities, and the standard of its science.  In all cases a healthy collaborative relationship is needed between the industry and service providers.

Page Contents

You cannot copy content of this web site. A copy protection system is operating.

New Zealand Walnut Industry Group