Research Knowledge and Needs for Orchard Floor Management in Organic Tree Fruit Systems

Abstract

While pest management and tree horticulture dominate the research agenda for temperate tree fruits, organic production requires more attention to weed control and nutrient management because of the limited options growers have. Weed control and tree nutrition are interlinked and can have impacts on other parts of the system such as rodent pests, beneficial insect habitat, water use, and soil quality. Trade-offs commonly occur when trying to solve weed control and nutrient management in organic orchards. This article reviews numerous orchard-floor management studies relevant to finding solutions for organic systems and suggests areas for further research. Alternatives to soil tillage are needed, as it can degrade soil and tree performance. Conversely, mulching frequently leads to improved tree performance and increased soil quality but can be a costly practice relative to other options. Currently effective organic analogues to the herbicide strip-grass alley system are lacking. A plant-based solution would be ideal from a sustainability standpoint, but no satisfactory solution has been developed for widespread use.

KEYWORDS:

• apple
• compost
• cover crop
• mulch
• nitrogen
• pear
• tillage

INTRODUCTION

The orchard floor represents a substantial portion of the orchard agroecosystem, but it has generally received less research and management attention than tree horticulture and pest management. Yet opportunities exist to improve orchard sustainability through manipulation of the orchard floor. Enhanced pest control and greater internal nutrient supply are two possible improvements that can be achieved, and these would improve the performance and lower the cost of organic tree fruit production. The desired outcomes of orchard-floor management are foremost focused on good tree performance (e.g., tree growth) and a high, consistent yield of quality fruit. Other outcomes are now receiving increased attention, such as biodiversity and soil carbon sequestration.

The role of the orchard floor is arguably more important in organic tree-fruit systems than in conventional orchards (Granatstein, 2003; Weibel and Häseli, 2003). Both weed control and tree fertilization can be achieved quite easily and with relatively low cost in conventional production using herbicides and synthetic fertilizers. This is not the case in organic systems, where poorly controlled weeds can compete with the trees for slowly released and expensive nutrient inputs. Organic systems generally substitute tillage for herbicides as part of the weed control program, but tillage is known to negatively impact tree growth and potentially degrade soil quality. Organic tree-fruit producers would welcome increased research on orchard-floor management that resulted in individual practices and their combinations that provided multiple benefits (e.g., weed control, fertility, carbon sequestration, insect biocontrol) to the orchard at lower overall cost.

Most orchard-floor management studies have been carried out in conventional orchards, but many of them have examined practices that would also be used by organic growers, such as tillage, mulching, and cover crops, and therefore are useful to consider. No single practice or set of practices is likely to perform equally well across the wide range of climatic, soil, and orchard conditions found where organic tree fruit is produced. Even the presence or absence of a single pest such as voles (Microtus pennsylvanicus) can dramatically shape organic orchard-floor management. Therefore, existing knowledge from the diverse growing regions needs to be examined for widely applicable practices as well as practices best suited to a more local set of conditions. The review presented here focuses on temperate tree-fruit experience and attempts to identify key considerations for organic orchard-floor management and suggest priority topics for future research.

Extent of Organic Tree Fruit Production

Organic tree-fruit production has been expanding in many commercial tree-fruit production regions in response to the continued increase in demand for organic fruit by consumers. Organic food sales have increased at 20% per year for over 10 years in the United States (Haumann, 2008; Sahota, 2008), with similar trends in Europe. Fruits and vegetables represent about half of organic food sales (OTA, 2008). Market penetration for organic fruit ranged from 5–7% of all fruit purchases in Switzerland, Austria, and Germany, according to recent data from BioAustria, AC Nielsen/ZMP/BOELW, and Bio Suisse. Over 70,000 hectares of temperate organic tree fruit were identified in 2006 that were located in all fruit-producing regions of the world (Granatstein et al., 2009a). Leading producers of selected organic tree fruit are listed in Table 1.

TABLE 1 Estimated World Area of Selected Organic Tree Fruits

	Hectares
	Apple	Pear	Cherry	Plum	Data year
Italy	2863	1420	2907	759	2006
Turkey	4278	2627	574	2471	2006
U.S.	7566	970	863	1195	2006
China	1600	1200	—	200	2005
Other N. Hemisphere	10,118	868	248	815
Argentina	1069	1354	—	400	2006
Chile	755	—	113	—	2006
Australia	150	120	50	75	2005
New Zealand	947	22	5	—	2007
TOTAL	29,356	8,581	4,760	5,915
Includes both certified and transition hectares. Transition area represents roughly 30% of the hectares reported in major producing countries of Italy, Turkey, and the United States. Adapted from Granatstein et al. (2009a).

Orchard Floor Components

The orchard floor can be viewed as a combination of five elements (soil, hydrologic cycle, nutrient cycle, flora and fauna, and energy) that interact with management. These elements have numerous intersections with orchard sustainability issues and the parts interact with one another:

• Soil – erosion control, compaction, nutrient cycling, gas exchange, soilborne disease control, carbon storage, pesticide decomposition

• Water and nutrients – intake and storage, minimal nitrate and chemical leaching, water conservation

• Biodiversity – potential to control pests, legumes to fix N, alternative crops

• Energy—frost protection, reduced energy inputs

The orchard floor generally consists of two zones with often distinct functions and management: the tree row, generally a 1–2 m strip in which competing vegetation is controlled, with the tree trunks in a line down the center; and the drive alley, typically planted to a perennial grass that is routinely mowed.

The orchard floor performs several important functions for an orchard system. These include:

• Physical support for tree growth, machinery operation

• Water intake/storage/transfer

• Nutrient cycling/storage (including litter decomposition)

• Gas exchange for roots

• Habitat (micro- and macrofauna)

• Microclimate (e.g., heat exchange)

These functions are impacted by various management practices, such as choice of understory species, tree architecture, irrigation system, nutrient inputs, spray drip, and weed control. In looking to design orchard-floor management systems for improved sustainability, trade-offs routinely occur. For example, while a vegetative cover under the trees might maximize soil quality, it will negatively affect tree performance. Therefore, continued research and grower experimentation are needed to test combinations of orchard-floor management practices that can optimize the system.

Limiting Competition with the Trees

A primary objective of orchard-floor management is to maximize tree growth and fruit yield by controlling the competition trees face from other vegetation. Competition for water and nutrients can come from weeds or planted cover crops. Fruit trees are relatively poor competitors with other vegetation, largely because of their low root density per unit of soil (Merwin, 2003). Apple (Malus x domestica) trees have 1–10 cm root per cm⁻² of soil surface, compared to 10³ cm/cm⁻² for grasses (Neilsen and Neilsen, 2003). Hogue and Neilsen (1987) identified four major orchard-floor management systems: permanent vegetation, mulching, cultivation, and herbicides. The standard system in high-density orchards is a vegetation-free strip (∼1.5 m wide) in the tree row managed with herbicides, and a perennial grass drive alley that is occasionally mowed. The “herbicide strip” or “weed strip” in the tree row eliminates competition while also reducing rodent habitat and maintaining water distribution by under-tree sprinklers. Merwin and Ray (1997) found that a 60–90 day weed-free period from May to July provided the best growth in apples. In their New York setting, a 2 m² weed-free area per tree, or a 0.7 m wide weed strip were the minimum needed. Weed growth influenced cumulative yield more than cumulative trunk growth, and tree nutrition was affected more by weed-free time than by weed-free area. Herbicide control of all orchard-floor vegetation often results in the greatest tree growth and fruit yield, but mulching systems combine good tree growth with positive effects on soil quality (Hogue and Neilsen, 1987).

Organic growers have several weed-control options from which to choose (Table 2). The most common management for weed control in organic orchards is tillage in the weed strip. It is relatively inexpensive but can negatively impact tree performance and soil quality. Another technique is thermal weed control that typically uses open propane-fired burners aimed at the weed strip and base of the tree. While low in cost, many weeds are not well controlled by flaming (Praat, 2002; Rifai et al., 1999). Several organic-compliant herbicides are now available, but they suffer from high cost, low effectiveness, and, in some cases, no crop-use label. All are contact postemergent materials. There is keen interest among some agriculture-supply manufacturers in developing a more robust organic herbicide (K. Treadway, personal communication), but certain organic standards might prohibit its use. Mulching using both organic materials such as wood chips as well as synthetic weed fabrics is also commonly used by growers. Continual mowing of the weed strip is another option that requires machinery that can work between the trunks. Organic growers often combine several of these techniques to achieve acceptable weed control as no one of them is fully satisfactory.

TABLE 2 Orchard Floor Weed Control Options for Organic Systems: Pros and Cons

	Pro	Con
Tillage	Effective	Reduced tree growth, fruit size
	Reduces rodent habitat	Costly in young orchards
	Relatively low cost	Can damage roots and trunks, irrigation system
		Can degrade soil quality, deplete organic matter
Flaming	Can control weeds around trunk	Potential tree injury
	Reduces rodent habitat	Not good for older weeds, perennials
	Relatively low cost	Uses fossil fuels
		Irrigation system damage
Inert mulches	Effective for most weeds	Costly to apply
	Can improve soil quality	Can tie up N
	Conserve moisture	May be hard to source
	Improved tree growth, yield
Living mulches	Add biodiversity	Compete with trees
	Benefit soil quality	Rodent habitat
	Legumes can fix N	Variable persistence
	Theoretically low maintenance	Variable ability to compete with weeds
Organic herbicides	Can control weeds around trunk	Expensive
	No physical damage to tree, roots	Inconsistent effectiveness
	Reduce rodent habitat	May need many applications
		Few registered products
Adapted from Granatstein and Mullinix (2008).

One approach that has been researched but not widely adopted is that of a “living mulch.” A species or mix of species of plants is established in the weed strip to fill the vacant niche that weeds are trying to fill. With ground cover and no disturbance, weeds can be greatly reduced (D. Granatstein, unpublished data). However, the challenge then shifts to controlling competition from the “living mulch” with the trees.

A New York apple trial compared eight different orchard-floor systems, including a crownvetch (Coronilla varia) living mulch, mowed sod grass, and herbicide-suppressed sod grass (2.5 m weed strip) over six years (Merwin et al., 1994). No supplemental irrigation was used, and moisture competition from the living mulches was severe, with the highest soil-water tension being recorded in those plots. However, the water holding capacity in the soil under suppressed grass (and mulch) ended up higher than under tillage or herbicide control due to a change in pore size distribution. The herbicide plots developed a surface crust and had lower water infiltration. The fewest tree roots were found under the living mulch treatments. The substantial reduction in tree shoot growth under living mulches and sod was linked to higher soil-water tension. Herbicide suppression of sod did reduce water consumption, but mowing did not. Mowed sod tended to be drier than suppressed sod but with more soil moisture than crownvetch treatments. Banded N fertilizer (166 kg N ha^-1) did not overcome the growth reduction by the living mulches. In contrast, cumulative fruit yield in 18-year-old tart-cherry (Prunus cerasus) trees in Michigan was not reduced by any of the three living mulch systems compared to the herbicide-strip control (Sanchez et al., 2003).

Living mulches, both legume and nonlegume, in a newly planted organic apple block in Washington State dramatically reduced tree growth and initial fruit yield. Increasing compost rate did not compensate for the competition perhaps because of the slow N release and/or the more competitive N uptake by the living mulch roots (D. Granatstein, unpublished data). Meyer et al. (1992) did find the greatest peach tree growth in North Carolina with bare ground or a living mulch of nimblewell (Muhlenbergia schreberii) perennial grass. A legume living mulch (full orchard floor) provided 120 kg N/ha over two years in a New Zealand apple trial, and equaled compost for tree leaf N. However, it also delayed fruit maturity and reduced firmness and color (Goh et al. 1995; Marsh et al., 1996). No competition-free control was included in the study for comparison. A living mulch using the weed horsetail (Equisetum arvense) led to virtual exclusion of weeds, and leaf N and tree vigor were the same as herbicide-treated plots (E. Hogue, unpublished data).

One compromise that has been developed is the “sandwich” system of weed control for organic orchards (Weibel et al., 2007). The concept is to put a perennial or self-seeding species in a narrow band in the trunk line, and then till each side of this to provide a competition-free zone for tree roots. The vegetation strip eliminates the more tedious work of weed control around trunks and reduces damage. In a Michigan apple trial, the “sandwich” system was the lowest cost practice for weed control, and when coupled with a rootstock adapted to it, tree performance and fruit yield were good (Stefanelli et al., 2009). In Washington State, the “sandwich” system had greater tree growth, but similar initial fruit yield compared to full living mulches, and lower yields than bare ground or wood-chip mulch (D. Granatstein, unpublished data).

A different approach was taken in an Argentine organic apple trial. Three different cover crops (all including a legume) were planted in the 3.2 m drive alley, while resident vegetation was left in the 0.8 m weed strip and mowed twice annually. The cover crops were mowed in place, and contrasted to a grower standard of disking the alley twice in late winter for late-frost passive control and letting resident vegetation grow back. The legumes generally led to increased trunk size and fruit yield compared to the control, but leaf N still declined over time (Sánchez et al., 2006). If the cover-crop clippings had been delivered to the tree row where most tree roots are located, perhaps this decline would have been prevented.

One common challenge among all attempts to fill the tree row with an acceptable living mulch is rodent control in growing regions with these pests. Voles (Microtus sp.) are particularly attracted by the habitat and can damage trees from their feeding. Measurement of vole presence in an orchard-floor management trial in Washington State found no difference between legume and nonlegume living mulches and no difference between comparable “sandwich” system treatments; all were much higher than control, tillage, and wood-chip mulch. However, a treatment with sweet woodruff (Galium odoratum) in a “sandwich” system had significantly lower vole presence than the other living mulches as measured by point intersects and surface runway lengths, but still more than the bare ground (Wiman et al., 2009).

Implications

• Fruit trees, especially dwarfing rootstock, are poor competitors with other plants for water and nutrients. Therefore competition must be limited to maximize fruit production.

• Filling the tree-row niche with a low-maintenance species or mixture that can exclude weeds but that has minimal competition with trees is a challenge. Strategies include suppression at key times or a cover crop growing period that avoids the main nutrient-uptake period for the trees.

• Any living mulch will need to be accompanied by a rodent-control strategy where rodent pests are present.

Suggested research

• Screening of novel plant materials for shallow root systems and low competition with trees.

• Screening of plant material for rodent repellency.

• Testing whether increased rootstock vigor can compensate for competition while not escalating tree management costs or whether differences in rootstock tolerance to competition exist.

Tillage

Tillage can provide effective weed control in organic orchards. New equipment is available to work between the trunks (e.g., Wonder Weeder® [Harris Mfg., Burbank, WA]; Ladurner mechanical hoe [Ladurner Co., Bolzano, Italy]) in a cost-effective manner with minimal tree damage. However, many research trials have found negative impacts of tillage, both on tree performance and soil quality (Hogue and Neilsen, 1987). This is particularly true on the coarse-textured soils of central Washington State where a large number of organic orchards are found. The U.S. Department of Agriculture National Organic Program (USDA, 2000) requires organic growers to “maintain or improve soil and water quality.” Therefore, alternatives to tillage are needed to keep growers in compliance.

In a 3-year Washington State organic-apple trial, tillage reduced tree growth and fruit size relative to the control, while mulching increased them. However, no clear negative impact on soil quality was measured (Granatstein et al., 2009b). In a subsequent trial with newly planted apples, tilled plots had among the best tree growth and yield (due to effective weed control), but experienced a decline in various soil quality indicators after two years (Hoagland et al,. 2008). Wooldridge and Harris (1989) found generally negative effects of tillage in a South African orchard, including a significant reduction in trunk size and pruning mass, attributed to damage to shallow feeder roots. Tillage also lowered soil cation exchange capacity (CEC) and available P, and led to a 13% drop in soil organic matter (SOM). While disking did not reduce SOM in an Argentine organic apple study, the cover-crop treatments experienced large increases in SOM (Sánchez et al., 2006). Tillage did lead to less tree vigor and reduced fruit bearing potential. The average SOM, N, exchangeable K and available P, Fe, and Zn were substantially greater in the surface soil (0–15 cm depth) due to the contribution of cover-crop biomass and leaf biomass from the apple trees. This stratification of soil nutrients emphasizes the importance of having good root-system distribution in the zone rich in available nutrients. Deciduous fruit-tree roots are typically abundant close to the surface where the soil is not disturbed by any type of cultivation (Hogue and Neilsen, 1987). Thus, damage to the root system from tillage may impact nutrient uptake by the trees. Caution should be taken when using tillage implements (e.g., disc, rotovator, Wonder Weeder®, Weed Badger® [Town & Country Research & Development, Marion, ND]) to control weeds. While the elimination of competing vegetation may benefit tree growth, the limitation of nutrient supply by root pruning may have the opposite effect.

Depth of tillage is typically 8–10 cm with available tools. No studies have examined depth of tillage to see if there is a minimal depth of disturbance at which tree performance is not impacted. If tillage is performed in the winter time when trees are dormant the negative effects are minimized, while the most damaging time appears to be around bloom (Ferree and Schupp, 2003). Therefore studies on timing of tillage are needed. Tillage studies have also not directly examined root damage and distribution. Neilsen et al. (1986) did find the majority of tree roots above 0.2 m depth with apples on M26 rootstock. Tools such as the Spedo blade (Rankin Equipment, Yakima, WA) or modified mowers or flails might be able to work at a more shallow depth and avoid root damage.

The negative impact of tillage on SOM can be managed by increasing inputs of carbon into the system. Several models exist (e.g. Soil Conditioning Index, USDA-NRCS) that can predict the loss due to various tillage equipment and the compensating organic amendments needed to maintain a steady state.

Implications

• Tillage for weed control can be effective and affordable.

• Tillage will generally lead to soil quality decline over time, driven by soil organic matter loss. Compensation through increased carbon inputs is possible.

• Tillage often leads to poorer tree performance, likely due to root pruning. It is not clear whether there is a threshold depth above which damage will not occur.

Suggested research

• Depth of tillage trial.

• Timing of tillage trial.

• Examination of root pruning and regrowth in tilled systems.

• Organic matter compensation; how much is needed, will this counter other soil quality declines.

• Contribution of the top soil to the nutrient supply of the trees.;

Mulching

Placing a mulch material in the tree row has consistently led to superior tree growth and often higher fruit yields in a number of locations and fruit crops. Many kinds of materials have been used as mulches: straw, hay, wood chips, paper, weed fabric, and even stones. The choice of material does influence tree performance to some degree, as it interacts with environmental factors such as temperature and moisture regimes. Mulches can influence soil temperature, soil moisture, and soil fauna; add nutrients; and suppress weeds. They can also provide habitat for pests (e.g., voles) and incite disease (e.g., Phytophthora root rot from excess humidity close to the tree trunk). Organic materials used as mulch, such as straw or wood chips, are bulky to handle and can be costly. Generally a material that has a higher C:N is preferable, as it lasts longer and is less hospitable to in-migrating weed seeds. All organic mulches will import some nutrients, and care needs to be taken to avoid creating imbalances, particularly for potassium. Inorganic mulches such as polypropylene weed fabric can last many years, provide excellent weed control and no nutrients, but can require significant labor to manage and are currently made from nonrenewable feedstock.

As early as the 1950s, Proebsting (1958) investigated straw mulching and other treatments in a peach (Prunus persica) orchard in the Yakima Valley, WA, with no limitation from water or nutrients other than N. The mulched trees had the most fruit, but not the greatest tree growth. In New York State, straw mulch and herbicides had the best apple tree-growth and fruit yield, tilled plots were intermediate, and grass or legume-sod plots were lowest (Merwin et al., 1994). After six years, tilled and herbicide plots held less water than mulch or suppressed grass treatments, and soil organic matter (SOM) increased in mulch plots and decreased in tilled and paraquat herbicide plots. Trees had the greatest number of roots under mulch, and the least under sod. It should be noted that root rots and voles were worse in mulch and crownvetch plots. Similarly in North Carolina, apple trees grew best in straw mulch and had the greatest fruit number (Shribbs and Skroch, 1986).

Apple-tree growth and fruit size increased with wood chip-mulch in a Washington State orchard trial relative to a control (Granatstein and Mullinix, 2008). One application of mulch provided weed control for 2–3 years, and voles were virtually absent in mulched plots. Irrigation water use was decreased by 20–25% under the mulch. In a subsequent trial with newly planted organic apples, trees again grew best with wood chip mulch, but the mulch degraded quickly each year and did not provide weed control that would be commercially acceptable (D. Granatstein, unpublished data).

Extensive research was done on mulches at Summerland, British Columbia, Canada on irrigated sandy soils with high density dwarf apple trees (Neilsen et al., 2003a). Shredded paper, biosolids, and compost were used alone and in combination along with alfalfa (Medicago sativa) hay, weed fabric, and bare ground control. Plots were fertigated with N beneath the mulches. All treatments exceeded the control for yield, including the weed fabric, by more than 50%. But the weed fabric trees were the same as the controls for trunk growth, giving weed fabric the highest yield efficiency. Alfalfa mulch led to larger fruit size, despite the fertigation with N. The researchers concluded that the mulch effects on tree growth were primarily nonnutritional.

Previous work by this group (Neilsen et al., 1986) had shown increased soil temperatures with black weed fabric in the tree row but higher fruit yield. Fabric treatment soil temperatures were 28–30^oC compared to 22^oC under sod. Gur et al. (1972) proposed an optimal soil temperature for apple roots of 18–25^oC. The temperature extremes lessened as the tree canopies grew and provided more shade. About two thirds of the tree roots were found between 0 to 0.2 m depth, where the greater temperature impacts would occur, and none below 0.9 m. Black weed fabric led to greater and earlier fruit yield than bare ground in sweet cherry at Hood River, Oregon (Tomasini et al., 2007) and more than paid for its additional cost of establishment. This is a cooler environment than Summerland and the additional soil warming was probably beneficial. More organic growers are using weed fabric because it is effective for weed control and has delivered improved tree performance. To avoid potential supraoptimal temperatures, some growers are removing it during the hottest time of year. New equipment for applying sheet material to orchards has been developed and can lower the cost. Fabrics may last 10–15 years in the orchard before needing replacement. New fabrics are being developed with white color on top (to reflect light and heat) and black on bottom to suppress weeds (Makus, 2007). Fabric may have a negative effect on some soil-quality factors such as water infiltration (Neilsen et al., 2003b) and soil fauna (Forge et al., 2003).

Forshey (1988) noted that mulches provide an ideal environment for trees, but that they are expensive and not enough material is available for their widespread use. He proposed the two options of a grass alley plus mulch around the trees or growing season cultivation with winter-cover crops. A survey of potential orchard-mulching materials in central Washington was done in 2002 (Granatstein et al., 2003) and found that enough material existed to potentially mulch 32,000 hectares of orchard (total orchard area exceeds 90,000 ha), but most materials had competing uses and would need to be purchased. One strategy to deal with this challenge is to develop systems that grow a mulch material in (or near) the orchard that can be handled mechanically and inexpensively. The “mow and blow” approach uses biomass growing in the drive alley as the feedstock and cuts it with an orchard mower that can deliver it to the tree row. Resident vegetation can be used, or species can be planted for a particular purpose (e.g., legumes for nitrogen fixation). The entire interrow space needs to be irrigated in order to maximize biomass production, and this may be limited by water supply and/or irrigation system design. Marsh et al. (1996) evaluated this strategy with apples in New Zealand and saw greater tree growth and increased fruit yield with the mulch. They cut a 3.5 m drive alley and delivered it to a 1 m tree row. A Washington State organic orchardist used a similar strategy, with alfalfa planted in the drive alley. But it produced excess N and encouraged voles. Simply using mow and blow with existing perennial grass in the drive alley led to fewer mowing trips, less herbicide use, less water use, and more N added to the tree row in another Washington State orchard that is now testing several planted legume species (using direct seeding) for their potential N contribution.

Implications

• Mulching generally leads to improved tree performance compared to other management systems. Nutritional and non-nutritional effects are involved.

• Mulching can be expensive and lower cost approaches are needed.

• Mulching needs to be tailored to the site to avoid potential problems.

• Presence of rodent pests will limit mulching options.

Suggested research

• Explore possible mulch effects on root systems and soil biota and their link to improved tree performance.

• Test various planted species for a “mow and blow” system, and evaluate timing and release of nutrients.

• Evaluate new mulch fabric options for their effect on trees, soil, and economics, including bio-based materials.

• Explore impact of different color mulches on tree performance and pests.

• Develop more rodent-control options.

Tree Nutrient Demand and Nutrient Supply

The orchard floor provides the base for tree nutrition in organic orchards. Nutrients are supplied by the soil through weathering of minerals, decomposition and recycling of plant residues, and storage and release of applied fertilizers and amendments. In Washington State, as in most producing regions with arid or semiarid climates, nitrogen is in limited supply in nearly all cases and is a key nutrient of concern in organic orchard-floor management. Large quantities of N are not exported in the fruit, but the tree-root system is not very efficient at utilizing applied fertilizer. The annual uptake of N for fruit trees can be approximated from the crop N content and assumptions about additional tree structure created in a given year (Sánchez et al., 1995). There are wide differences in annual requirements for various tree species. Apples, pears (Pyrus communis), and cherries (Prunus avium) remove less than 40 kg N ha⁻¹ yr⁻¹, some nut crops remove over 100 kg N ha⁻¹ yr⁻¹, and most stone fruit remove intermediate amounts (Weinbaum et al., 1992).

Neilsen and Neilsen (2003) cite N removal (kg ha⁻¹) from orchards in fruit and dead leaves as follows: end of year one: 9.4; end of year three: 34.7; end of year six: 40.2. If leaves and prunings can be retained and the N recycled in the orchard, only the 30 kg N ha⁻¹ that is actually leaving in the fruit each year at full bearing needs to be replaced. Of course, nitrogen is accumulated in the growing trees as the orchard is established, starting at 7 kg N ha⁻¹ in the nursery trees themselves, rising to 27 kg N ha⁻¹ at the end of the planting year, and to 65 kg N ha⁻¹ at the end of the fourth year in a dwarfing, high-density orchard. The trees store a considerable amount of N in the trunk, branches, and roots, and this is enough to supply as much as 40% of the growing season needs in mature trees (Sánchez et al., 1995).

The annual N requirement for high-yielding ‘Comice’ pear trees in Oregon was estimated to be 48 kg N ha⁻¹. Of that, 18.5 kg of N corresponded to the fruit requirement. Nearly 45% of total N demand was met by tree reserves (Sánchez et al., 1991). An apple crop in New Zealand was estimated to remove 58 kg N ha⁻¹ in the fruit and from leaching loss (Goh et al., 1995), while a clover cover crop provided nearly that much N in its biomass, illustrating the potential for organic orchards to supply a significant portion of their N needs internally. Most of the cited nitrogen estimates were obtained from conventional orchard studies that examined N partitioning among different plant parts (fruit, leaves, branches, trunk, and roots). The use of ¹⁵N tracers allowed researchers to study the seasonal uptake patterns of those parts. These basic studies are useful for organic orchards because they identify the key timing and location of plant N demand that needs to be met with N released from organic fertilizers and/or cover crops. Thus several strategies can be developed to maximize N uptake and minimize possible N losses from the orchard (leaching, runoff, etc.).

Management of nutrients in an organic system is complicated by the slower and less predictable mineralization of the organic materials in which the nutrients are embedded. This process is mediated by the soil fauna, which are impacted by carbon (energy) availability, temperature, and water. Providing nutrients at a precise time in organic systems is therefor more challenging. For fruit trees, availability of high amounts of nitrogen at the wrong time can have negative impacts on fruit quality by delaying maturity, or reducing firmness and color (Granatstein and Mullinix, 2008). This occurred in New Zealand apples with a “mow and blow” system using red clover (Trifolium pratense). Soil N and leaf N were both increased along with tree growth and fruit yield, but fruit quality was lower (Marsh et al., 1996). The clover released high amounts of N midseason when it negatively impacted fruit quality.

Soil nitrate levels up to 130 mg kg⁻¹ soil were measured in November and December (equivalent to May and June in the Northern Hemisphere) in an Argentine organic-apple orchard in the top 20 cm of soil, and 40 mg kg⁻¹ soil in the 40–60 cm depth soil layer because of natural decomposition of the entire vetch (Vicia sativa) cover-crop biomass (Sánchez et al., 2006). The low C:N ratio of the vetch, along with high temperatures and consistent irrigation, favored N mineralization over a few weeks. The sharp peak of nitrate release was correlated with a flush of tree growth, which is not desirable because it may cause physiological disorders in the fruit and transient nutrient deficiencies (Faust, 1989). A more desirable management of the cover crop involves multiple cuts of the legume during the growing season to deliver smaller amounts of N, avoiding decomposition of the entire biomass at one time.

Thus, nitrogen from legume crops in the orchard can be transferred to the trees. In Chile, N was transferred from clover to raspberry (Rubus idaeus), and 18–22% of the N in the leaves came from the legume as measured with ¹⁵N dilution. Raspberry biomass and fruit yield were higher in the cover crop treatment (Ovalle et al., 2007). Apple tree-leaf N was significantly higher with unfertilized white clover (Trifolium repens) living mulch compared to a fertilized bare-ground control one and two years after clover establishment, but the clover also led to slower leaf senescence in October due to late season N (Granatstein and Mullinix, 2008). Neilsen et al. (2003a) found leaf P in apple trees to be more affected by mulch and organic amendment treatment than leaf N. Excess leaf K depressed leaf Mg, but there was no treatment effect on leaf Ca. Mulches tended to increase leaf Zn and decrease leaf Cu. In an earlier apple study, they found that grass sod increased fruit K concentration, which reduced fruit Ca. Leaf N and trunk size were lower, fruit firmness was consistently greater at harvest and out of storage, but fruit size was smaller. Adequate leaf N was achieved with 30 kg N ha⁻¹ yr⁻¹ with no sod (Neilsen et al., 1984). In New York apples, the effect of N fixation by crownvetch living mulch was not observed in the trees, and a banded application of 66 kg N ha⁻¹ did not overcome the growth reduction by any of the living mulches (Merwin and Stiles, 1994).

Potential nutrient imbalances need to be considered by organic orchardists. Nutrient sources from animal manure, a common input, are typically enriched in phosphorus relative to the original plant materials, and this is often exacerbated by the composting process, which loses mass of C and N but not P and K. For example, to get 50 kg of plant-available N (in the season of application) by using alfalfa you also get 10 kg P and 80 kg K per hectare. With chicken manure compost, you get 26 kg P and 39 kg K per hectare. Higher analysis materials such as feathermeal or bloodmeal (∼12% N) are available but are considerably more expensive per kg of N. Yet they mineralize faster and over a shorter period of time, allowing for more precise nutrient management. An organic fertility calculator has been developed by Oregon State University that is a very helpful tool for assessing nutrient choices in organic systems (http://smallfarms.oregonstate.edu/organic-fertilizer-calculator). New processes are being investigated to extract and concentrate N and P from organic wastes, but it is unclear which of these will lead to organic compliant products.

If more nutrients can be supplied and recycled within the orchard, some of the above concerns could be mitigated. By growing a legume in the orchard, if it is not being fertilized, it will bring in new N through atmospheric fixation as needed and utilize existing P, K, and other nutrients for growth. When it is mowed, these nutrients are recycled and used for the next growth of the legume. If a “mow and blow” strategy is used, then there is potential for enrichment of the tree row with P and K, and depletion to the point of nutritional need in the drive alley. Wooldridge and Harris (1989) did see some excess K in leaves and fruit with a “mow and blow” system in apples. So an ideal design would grow the legume in the tree row where it primarily adds new N. Yet as discussed above, this will almost certainly lead to unacceptable competition with the tree. If the legume could grow during a period when the tree was not actively taking up nutrients, competition could be reduced. This is possible in climates where winters are mild enough to either grow a cover crop, or to utilize late fall and early spring for enough growth to produce substantial biomass. In central Washington State, the climate is not well suited to this strategy. Trials with cover crops generally have shown that plantings made later than mid-September do not lead to substantial biomass accumulation before the low temperatures stop growth. With warmer temperatures coming earlier in spring, there may be an opportunity, and this could be combined with a dormant seeding so the seed would already be in place. Most likely an annual species would fit best.

Many other potential strategies remain to be explored. These include combining a spring organic fertilizer with a late summer “mow and blow”; a winter annual legume for spring N and a perennial legume for late summer N, direct seeding the annual legume into the perennial; and a tree-row cereal cover crop to absorb any excess soil N in late summer and store it until the spring. In addition, the soil biota may be able to be managed to store nutrients during certain periods and release them at others. Various mulches and amendments had large effects on the soil food web and on N and P cycling in a British Columbia apple orchard (Forge et al., 2003).

Implications

• Organic systems can supply adequate nutrients to the trees.

• Timing of supply is a challenge, but slow release from organic nutrients may be beneficial.

• Growers need to consider potential nutrient imbalances from continual use of certain organic inputs.

• Use of carefully designed cover-crop strategies within the system has the potential to add N without creating nutrient imbalances. Improved rodent control is a prerequisite for their use.

Suggested research

• Matching nutrient needs and nutrient release from organic sources.

• Screening of legumes as a nitrogen source in a “mow and blow” system.

• Evaluation of fall-planted legumes for a spring/summer nitrogen input.

• Evaluation of a fall catch crop in the tree row to scavenge N, improve fruit quality, and release N in the spring.

• Determination of effect of slow-release nutrients from organic fertilizers on tree performance relative to soluble or quickly available fertilizers.

• Continued research on higher analysis nitrogen sources compliant with organic production, especially recycled from existing organic wastes.

Soil Quality

Effects of orchard-floor management on soil quality have been touched on in the preceding sections. Clearly, organic inputs tend to improve soil quality while tillage and bare ground can degrade it. Over five years in a New York apple-orchard trial, soil organic matter increased with mulch; was unchanged with living mulch, mowed and suppressed sod, and glyphosate herbicide; and decreased with tillage and paraquat herbicide treatments (Merwin et al., 1994). Various organic amendments increased microfauna in a British Columbia apple trial compared to bare-ground control or weed fabric. Combinations of paper mulch and biosolids elevated bacterial feeding and omnivorous/predatory nematode and reduced root lesion nematode (Pratylenchus penetrans) numbers below the damage threshold (Forge et al., 2003). In Michigan tart cherries, living mulch, “mow and blow,” and compost all generally increased total nematode numbers, but did not lead to consistent reduction in root lesion nematode (Sanchez et al., 2003). Reich (1985) reported a reduction in mycorrhizal infection of roots in potted apple trees with grass.

Living mulches (both legume and nonlegume) and a brassica seed meal amendment significantly increased total nematode numbers over two years in a new Washington organic apple planting, and the living mulches also had somewhat higher carbon mineralization and dehyrdogenase activity in the soil, all suggesting enhanced soil biota (Hoagland et al., 2008). Despite inputs of compost, tilled plots had the lowest soil C and N after two years.

In a side-by-side apple systems comparison in Washington State, both organic and integrated systems led to a higher soil quality index rating than the conventional system over five years (Glover et al., 1998). Under the function “resist degradation,” the organic soil was rated the same as conventional and lower than integrated, probably due to tillage in the organic but not in the integrated treatment. Wooldridge and Harris (1989) reported a 13% drop in SOM with tillage compared to the herbicide control along with reduced CEC and lower available P.

Orchard floor management influenced nitrate leaching from the soil in both a New York apple and a Michigan cherry study. A full orchard-floor cover crop reduced leaching by >90% in tart cherry compared to the conventional herbicide-strip system, and compost as an N source had a similar effect (Sanchez et al., 2003). There was less pesticide and nitrate loss in apple with mowed sodgrass or biomass mulch, while postemergent herbicide treatments had the most (Merwin et al., 1996).

In a Michigan trial using 18-year-old tart-cherry trees, a living mulch and “mow and blow” mulch with and without composted manure increased total C and the active C and N pools in the soil. “Mow and blow” increased soil C by >20% compared to the herbicide-strip control. The soil active N pool was increased 25% with living mulch or “mow and blow,” and 60% with compost, enhancing the overall N-supplying capacity of the soil (Sanchez et al., 2003). Similar soil benefits were achieved in a New York apple study, but orchard-floor management did influence tree growth and fruit yield. Wood-chip mulch or mowed-grass sod generally had higher levels of soil microbial activity and populations than either of two herbicide-strip treatments (Yao et al., 2005). Mulch had the highest soil respiration, SOM, CEC, and available P and Ca, as well as a different fungal community than the other treatments. However, mulch tended to have larger tree growth and fruit yield, while grass had the lowest.

These studies illustrate the difficulty in extrapolating measures of soil quality to predictions of tree performance. Most soil measures are from bulk soil and not the rhizosphere of the tree roots. It is not clear how treatment contrasts would compare under these two different sampling regimes. Changes in several soil quality factors, including water intake and storage, soil organic matter, or potential N mineralization, would be expected to promote better tree performance, and can lead to other positive sustainability outcomes (carbon sequestration, less nitrate leaching). The direct role of soil biota in influencing tree performance could stem from growth-promoting compounds, exclusion of pathogens, nutrient or water transfer (mycorrhizae), pathogenicity or root feeding. More research is needed to establish whether there is a clear beneficial relationship between certain soil biological conditions and tree performance.

Implications

• Living cover crops and organic inputs tend to enhance soil carbon, nitrogen, and microbial status.

• Tillage and continually bare soil tend to lead to decline in soil carbon and related parameters.

• Management systems can significantly change the nematode community of the soil, with potential impacts on nutrient cycling and plant-parasitic nematodes.

• Soil quality improvements are not always associated with improvements in tree performance, but they represent changes that may be more highly valued in the future.

Suggested research

• Comparative studies of soil biology under different orchard-floor management that contrast bulk soil and rhizosphere sampling.

• Role of orchard-floor management in mycorhizal colonization of tree roots.

• Role of orchard-floor management in stimulating free-living nitrogen fixers in soil, and the potential to introduce and sustain these organisms.

Pests

The orchard floor represents a large habitat opportunity for many organisms. Growers already avoid certain practices known to favor organisms that are pests of fruit trees, such as bare, dusty soil that incites mite outbreaks and legume cover crops that harbor insect pests such as Lygus species. The idea of enhancing biocontrol of orchard pests through conscious choice of orchard-floor vegetation and management has also been explored, with few examples of clear success. Attempts to increase generalist natural enemies may work, but these organisms often do not exert control of pests in the tree canopy. For example, reducing mowing frequency in Washington pear orchards led to higher numbers of beneficial insects in the orchard-floor vegetation but also dramatically increased stink bug (Euschistus conspersus and Acrosternum hilare) numbers (a pest species) and did not lead to measurable biocontrol in the trees (Horton et al., 2003). Similarly, attempts to increase orchard-floor floral diversity did increase insect biodiversity but did not augment biocontrol of apple pests in California (Caprile et al., 1994). Meyer et al. (1992) found that a number of orchard-floor ground covers in peach incited fruit pests in their North Carolina trial.

There is growing evidence that general enhancement of biodiversity does have positive effects on biological control (M.W. Brown, personal communication; Snyder et al., 2008), despite the lack of success in several orchard studies. Fernández et al. (2008) studied the arthropod diversity in apples under five different orchard-floor management regimes in northern Patagonia, Argentina. Overall, 53% of the collected species were phytophagous and 42% were biological controllers (BC). The alfalfa-fescue (Festuca arundinacea) cover crop had 60% BC, and strawberry clover (Trifolium fragiferum) had 56% BC. Both treatments also had the highest diversity. In comparison, the native-vegetation check and the conventional check had the lowest indices of diversity. Most arthropod pest levels were determined by the pest control activities. The woolly apple aphid (Eriosoma lanigerum) required no control in the organic blocks, while control was necessary in the conventional block.

Successful examples usually involve a specific pest-predator relation. In Georgia pecans (Carya illinoinensis), the yellow pecan aphid (Monelliopsis pecanis) was controlled with convergent lady beetle (Hippodamia convergens) by growing a cover crop of hairy vetch (Vicia villosa). This produced two generations of lady beetles, reaching 353,000 per hectare, and they migrated from the senescing ground cover to the pecan trees at the time when aphids were reaching peak levels (Tedders, 1983). In China, natural enemies (Amblyseius sp.) of the citrus red mite (Panonychus citri) were encouraged on the weed Ageratum conyzoides. By conserving or planting this species in the citrus groves, biocontrol was achieved, and the practice was used on over 135,000 hectares of citrus (Liang and Huang, 1994). The planting of wild rose (Rosa woodsii) hedges around apple blocks in Washington State helped retain the leafroller parasitoid Colpoclypeus florus that fed on the strawberry leafroller (Ancylis comptana) in the roses, and then migrated into the orchard when fruit leafroller populations began to build (Pfannenstiel and Unruh, 2003). Current research in Washington State is examining the potential for alfalfa to host pear psylla (Cacopsylla pyricola) natural enemies, and testing several plant species for the ability of their flowers to host specific natural enemies of woolly apply aphid.

Orchard-floor management can influence pests indirectly through regulation of N and water that affect tree vigor. There was a much-reduced aphid infestation in British Columbia apples with a white clover/grass living mulch versus a herbicide strip, weed fabric, or winter rye (Secale cerale) cover crop (Haley and Hogue, 1990). The clover mix competed with trees, reducing their vigor, which reduced the aphids. A similar effect was seen in West Virginia apples comparing a living mulch with a full herbicide strip (Brown and Schmitt, 1996). Diseases, especially fireblight (Erwinia amylovora), were also less problematic in the trees with the reduced vigor.

Use of entomopathogenic nematodes (EPN) in orchards to control codling-moth larvae has had limited success due to the need for high humidity for nematode survival while they search for prey. Wood-chip mulch in the tree row provided a more favorable environment, and up to 95% mortality of codling-moth larvae was achieved with Steinernema carpocapse and S. feltiae (Lacey et al., 2006). More research is needed on the location of the overwintering larvae and possible ways to attract them to a habitat like mulch where the nematodes can exert high mortality on this part of the codling moth life cycle that typically has not been addressed. A compost application under apple trees in West Virginia significantly affected arthropod abundance during two years after application, with more predators and fewer herbivores in the compost treated plots (Brown and Tworkoski, 2004). Populations of spotted tentiform leafminer (Phyllonorycter blancardella) and migrating woolly apple aphid (Eriosoma lanigerum) nymphs were reduced in the compost plots.

As cited above, orchard-floor management can impact pathology, with regard to straw mulch and diseases such as Phytophthora (Merwin and Stiles, 1994). Choice of ground-cover species in a California pear orchard influenced fruit finish, with packouts greater under legumes and herbicide-treated vegetation than under annual ryegrass, perennial ryegrass, and resident vegetation (McGourty, 2009). Placing plant residue, such as “mow and blow,” under the tree row can increase the risk of fruit infection by saprophytic fungi such as Penicillium and Botrytis and impact postharvest fruit quality (C.L. Xiao, personal communication), an outcome experienced by a Washington State pear grower. Apple-replant disease has been suppressed by orchard-floor strategies such as a wheat (Triticum aesetivum) cover crop, overwinter trenching of the tree row, and brassica seed meal amendments to the tree row (Mazzola et al., 2002). Compost application in California soft fruit orchards was linked to reduction in brown rot (Monolinia fructicola) on the fruit (Daane et al., 1995). The influence of orchard-floor management on disease will depend on such factors as the climate, irrigation system, fruit type, soil type, canopy architecture, and light penetration.

Rodent pests (if present) are influenced by orchard-floor management, as discussed above. Whereas conventional growers can deploy rodenticides to control populations, similar materials are not available for organic production. A few organic orchardists have tried mass trapping with mechanical mousetraps and reported successful results at an acceptable cost (M. Robinson, personal communication). But many of the potential orchard-floor management strategies discussed above that might benefit tree growth, nutrition, weed control, or soil quality will definitely increase the risk of rodent damage.

Implications

• Orchard-floor vegetation management can clearly affect pests and natural enemies in the orchard; however this does not always translate to successful control.

• The most success has been achieved when a specific plant-natural enemy-pest combination can be identified.

• Amendments such as mulch and compost may offer pest management opportunities in addition to their soil benefits.

• Rodent pest damage potential is increased by many alternative orchard-floor management practices.

Suggested research

• Continued studies of orchard-floor vegetation and associated fauna and their interaction with fruit pests using techniques such as protein marking.

• Research on attracting overwintering codling moth to orchard-floor locations where EPNs can exert significant mortality.

• Evaluation of the pest and disease impacts of commonly used soil amendments in organic orchards, such as compost.

• Research on the timing of mowing to encourage natural-enemy movement to the trees.

• Research on alternate-row management of vegetation composition, mowing timing, type of mower, and other practices that could influence biocontrol.

• Test introduction of poultry or other economic animal species that may increase predation of pests.

• Development of new control options for rodent pests, especially voles.

SUMMARY

Orchard-floor management is a more critical and challenging pursuit in organic tree fruit because of the limited options growers have for controlling weeds and providing adequate amounts of nutrients at the appropriate time. Practices that may help solve these problems, such as mulching, often lead to an elevated risk of tree damage from rodent pests, for which there are few control options. The orchard floor also may offer an opportunity to enhance insect biocontrol, another benefit that can help offset the reduced number and efficacy of pesticides allowed in organic systems. In order to balance the sometimes competing needs for weed control, soil quality, and tree nutrition, more research and innovation are needed to identify plants or plant mixtures that can perform multiple functions as part of the orchard floor. The utility of such an approach is dependent on available water for their growth. Regions with insufficient irrigation water to wet the entire orchard floor may not be able to consider plant-based solutions. Additionally, orchards with irrigation systems that limit water delivery to the tree row (e.g., drip irrigation) will face the same constraint. Other technologies that avoid soil tillage, such as weed fabrics, must also be considered. A plant-based solution to orchard-floor management problems will not likely fit all regions and orchards, but would contribute to improved sustainability in organic systems. Nutrient recycling, biological pest and disease control, and high-quality fruit are common goals for organic-fruit growers worldwide and are all influenced by orchard-floor management. The constraints on organic systems often require more local adaptation of technologies than in conventional systems, leading to limited responses to climatic and pest factors that shape the development of orchard-floor management. Currently no analogue to the more universal “grass alley-herbicide strip” system used in conventional orchards exists for organic systems, necessitating continued research and field experience to improve organic orchard-floor management.