Best management practices for organic orchard nutrition

Final Report for LS05-176

Project Type: Research and Education
Funds awarded in 2005: $200,000.00
Projected End Date: 12/31/2008
Region: Southern
State: Arkansas
Principal Investigator:
Dr. Curt Rom
University of Arkansas
Expand All

Project Information


Project Summary - Development of an Best Practices for Organic Orchards for the South
The development of best practices for sustainable organic orchards in the southern region was addressed through research studies: surveys of current grower practices and nutritional status, and a replicated field trial studying the interaction of orchard nutrient sources and ground cover management. The grower surveys indicated a wide range and no consistency or standards in nutrition management for organic managed orchards. Nutritional levels of orchards tended to be at the low range or in deficiency compared to conventional standards. In the replicated trial, it was clear that ground cover management system was more important than nutrient source during establishment. Trees grown with a mulch of green compost or woodchips applied annually grew sufficiently to fill allotted space and bear a crop in the third year, while trees with a mow-and-blow mulch or shredded paper mulch, did not. Nutrient source did not have as significant effect on tree growth but did have significant impacts on cropping potential in the third growing season.

Project Objectives:

Objectives/Performance Targets
The goal of this study is to identify, test and/or develop best management practices for organic orchard nutrition and ground cover management during establishment with the following specific objectives.

A. Identify successful nutrient management practices currently being used by organic fruit growers in the South

B. Work with Southern region organic tree fruit growers to assess the impact of their nutrient management practices on soil quality, tree nutrient content, fruit yield and quality, pest incidence, and management costs

C. Based on results of on-farm analyses, conduct a controlled, replicated study to further evaluate ground cover and nutrient management practices and develop locally-appropriate recommendations for organic fruit tree growers in the South.

D. Assist tree fruit growers implement effective organic nutrient management practices by working with grower collaborators, grower organizations, Cooperative Extension, and ATTRA to conduct on-farm trainings and develop informational materials


Click linked name(s) to expand/collapse or show everyone's info
  • M. Elena Garcia
  • Donn Johnson
  • David Miller
  • Anthony Owens
  • Marvin Owings
  • Jennie Popp
  • Mary Savin


Research results and discussion:

Grower Surveys.

Of the surveys mailed, 17% were returned and 12% were completed in sufficient detail to be useful. There were no certified organic orchard producers that responded, but all respondents used organic practices. The vast majority of the producers (85%) indicated orchard size was increasing and three indicated their orchards were new and not yet producing. All of the producers stated that nutrition was important to their orchard productivity and 85% conducted bi-annual or annual soil testing. Few producers, however, used foliar nutrient analyses as a routine component of their management regime. The most common symptom of nutrient deficiency reported was small fruit size (42%). The most common types of nutrients applied included composted mulches (71%), composted manures (43%), rock minerals of phosphate (43%) and calcium (43%), boron (43%) and magnesium sulfate (43%).

As a result of the apple producer survey, two apple producers, employing organic practices, were identified and agreed to monitor soil and foliar nutrients in their orchards. In the summer of 2006, producers submit two soil and foliar samples. Growers were sent protocols and instructions as well as prepaid return mail envelops. After analyses of the soil and the foliar samples, and after review of the grower initial surveys, recommendations were made by UA pomologists researching organic production. In both cases, the foliar samples were in the low range of N and Mg, barely adequate. One grower had deficiency levels of Ca and B. In both cases grower samples were in the adequate to high range for foliar P and K (Garcia, 2007; Shear and Faust, 1980).

Both growers agree to participate in 2007, however a record breaking frost event in April 2007 caused all fruit flowers to freeze in the southern region preventing any harvest. As a result neither growers were interested in the testing program.
Nutrient contents of ground cover mulches and nutrient sources. The mulches varied in [N], C/N, and provided a range of nutrients to the system (Table 1). Likewise, the nutrient sources of the PL and CF provided a range of nutrition. For nutrient source treatments, PL and CF were applied to rates to equalize the amount of N applied per tree per year. Roughly, PL was applied at a volume rate of 5.8-times the CF to equalize the [N] applied. The SP mulch had very low [N] by but high [C] and a resulting very high C:N ratio. The GC treatment applied total N at a rate similar to PL.

Tree Growth
Trees grown with WC or GC mulches were tallest and had the greatest TCSA in both year, and had the greatest increase during the establishment period of the study (Table 3). Although trees with MB were smallest after the first year of growth, they increased significantly in size and were slightly larger than trees with the SP mulch. There was no effect of nutrient source on tree height in both years and TCSA in the first year. However, in the second year of growth, trees receiving CF had the largest TCSA.

After two seasons of growth, trees with WC and GC had achieved the height of the training system (>10.5 ft or 3.25m) and were of sufficient size to crop in the third season, 2008. All trees were allowed to bloom in 2008. Trees with the GC and WC mulches had significantly higher flower and fruit number per tree than those receiving MB or SP (Table 4). Trees with the SP treatments had significantly lower fruit set. Trees receiving nutrients from the PL or CF sources had significantly more flowers and fruit per tree than the nonfertilized control (NF) and slightly reduced fruit set.

Therefore, although nutrient source did not appear to affect tree growth and development significantly, the lack of supplied nutrition (NF treatment) resulted in significantly reduced production potential in the first cropping season.

Similarly, trees with WC had the largest canopy volume and total leaf area per tree, canopy leaf density, followed by GC while trees receiving MB and SP had significantly less canopy volume and leaf area (data not presented). Leaf area index, an indicator of physiological efficiency of the canopy was greatest in the GC and lowest in the SP treatments. Trees with WC and GC had the largest average leaf are and trees with WC had the greatest specific leaf weight (SLW), an indicator of leaf physiological activity. Estimated leaf chlorophyll (CHL) was significantly lower in SP trees in 2007 than all other treatments, and trees with WC and GC had the highest estimated chlorophyll contents. Midday photosynthesis in the second season correlated to SLW and CHL.

Physiological basis for growth and cropping potential can be found in both the nutritional status of the trees, and the soil nutrient status, quality, and biological activity.

Tree Nutrition
During the first two growing seasons, the GC and WC had the highest foliar [N] while the trees with SP had significantly less [N], and had deficiency levels (Tables 5 and 6). In the first year, trees provided with the control NF nutrient source had significantly less and deficient levels of foliar [N]. In the first year, although there were some differences among treatments for some nutrients (e.g. K, S, Mn), all nutrients were in a sufficiency range (Garcia, 2007; Shear and Faust, 1980). In the second season there were also significant differences in foliar nutrient contents (Table 6) although most nutrient concentrations were within an adequate range. All ground cover treatment and nutrient source treatments resulted in adequate ranges of foliar [P], [K], [Ca], [Mn], and [B] but in the low range or deficient for foliar [Mg], [S], and [Cu]. The complexities of the effects of treatment on leaf size confound some of the observed differences such as the small leaf size observed with SP treated trees was negatively related to foliar [P], [K], and [Ca]. This was also observed for the NF nutrient source treatment which had the highest foliar [N], [P], [K], [Na], [Fe], [Mn], [Zn], [Cu], and [B] contents but the smallest average leaf size, SLW, and leaf area per tree.

For the ground cover management systems, it appears that foliar [N] was related to the growth and cropping potential of the trees.

Soil Quality and Nutrition
In the first growing season, there were no differences among ground cover treatments for soil bulk density or water infiltration. In the second growing season, the soil bulk density in the top 10 cm soil was lowest in the MB treated plots during October (data not presented), but SP treated plots had the highest soil bulk density (e.g. heavier, more compacted soils).

Soil pH, Organic Matter, and Electrical Conductivity
Since the preplant condition (Table 1), the addition of preplant lime and treatments has significantly increased pH from approximately 5.2 to an average of 6.8 in all test plots. The ground cover management systems had an impact on soil pH, organic matter content (OM) and electrical conductivity (EC) or salt content. The soil pH in the top 10 cm soil was greatest in the SP plots throughout the second season (data not presented). Plots treated with GC tended to decrease in pH during the season compared to other treatments. Soil pH of all nutrient source treatments decreased from March until July then slowly increased for the remainder of the growing season. In May, plots treated with PL and CF had lower pH than the NF treatments, but pH was similar among all treatments in September. Correspondingly, soil EC, an indication of dissolved salt ions, increased early in the season with GC plots having significantly higher EC. The NF nutrient source treatments generally had lower soil EC than other treatments. All ground cover treatments appeared to increase soil organic matter from preplant levels (~1.0% to ~1.5%), but the soils were still relatively low levels. GC treated plots had higher soil OM than other treatments early in the season, but there was no difference among treatments as the season progressed. At 10-30cm, the OM content tended to increase during the season, even though OM content was low approximately 1%. There were no significant differences among nutrient treatments on soil OM.

Soil Nutrient Contents
Ground covers management significantly increased soil [NO3-N] from preplant levels, with the GC treatment having the highest soil [NO3-N] (Table 7). The ground cover treatment increased soil [K], [Mg], and [Na] compared to preplant levels. GC treatments had the highest soil [K], [S], and [B] in both the first and second years. The SP treatments had the lowest soil [NO3-N], [P], [K], [Mg], and [B] in the second year.

The nutrient sources also affected soil nutrient content (Table 7b). The NF treated trees were lowest in soil [NO3-N], [P], [K], [Zn], [Cu], and [B] compared to CF or PL trees. CF treated trees tended to have the highest soil [NO3-N], and [K] while the PL trees had highest [Zn], and [Cu] .

Soil Biology
Ground cover management systems significantly affected soil biological activity (Tables 8 and 9). Due to unforeseen circumstances and harsh environmental conditions, soil samples for microbial biomass and enzymes were not collected in spring 2006. Data in 2007 suggest that those initial ground cover treatments had immediate impacts on soil microbial biomass, available nutrient pools, and enzyme activities, especially in the surface 10 cm, and that those impacts persisted into 2007. Ground cover applications then had an impact in 2007, as values for many properties, and differences among ground covers, changed after applications in 2007, again more so in the surface 0 - 10 cm, but also in some properties at the 10 - 30 cm depth. The ground cover x depth interaction was significant before applications in March and after applications in May

The three-way interaction of ground cover x fertilizer x sampling time (before and after treatment applications) has not yet been analyzed. The three-way interaction of ground cover x fertilizer x depth was not significant at P < 0.05, although it would be in some cases at P < 0.10 (data not shown). Sampling time was a significant main effect for all properties except ammonium (data not shown). For all properties where sampling time was significant, values were lower after treatment applications except for microbial biomass C:N ratios, nitrate and dissolved total N. Fertilizer was rarely significant, except for available N pools after applications (data not shown).

Ground cover x depth was frequently a significant interaction. The interaction was not significant for microbial biomass N, β-glucosaminidase, and ammonium in March (before applications), and it was not significant for microbial biomass C:N ratios and ammonium after applications (data not shown). The only instance for which depth was not significant was for microbial biomass C:N ratios after applications (no significant differences, data not shown). The importance of depth to results is not surprising as microbial biomass, available nutrients, and enzyme activities are expected to decrease with depth.

Data analysis is on-going; however, these preliminary analyses show that ground cover treatments in this transitional orchard are impacting microbial biomass, available nutrients, and enzyme activities. These properties are being measured because they are good, responsive indicators of short-term changes in soil quality and functional qualities involving soil nutrient dynamics. These surface treatments are impacting the top 10 cm of soil and, to a lesser extent, further into the root zone down to 30 cm in the profile. These properties were analyzed in conjunction with tree responses which are also already showing different responses to ground covers as well.

Soil respiration (SRs) was monitored monthly throughout 2007. Mid-season SRs was highest in the GC treatments, intermediate in WC and MB treatments, and lowest in the SP treatments (data not presented). There was no difference in SRs among nutrient source treatments across the mulch treatments.

Although not significantly different due to significant variation and limited sample sizes, apple tree root density was highest in the GC and WC treatments and least (80% reduction) in the MB treatment (data not presented). There was no difference among nutrient source treatments for root density.

Soil Temperature and Moisture
Soil temperatures tended to be higher in the MB treated plots and was consistently lowest in the SP plots (data not presented). Soil temperatures with GC and WC were generally lower than the MB but higher than the SP plots. There was no effect of nutrient source on soil temperatures.

Soil moisture was generally kept within an acceptable, nonstress range between 10 and 50cbars soil water tension (Tables 10 and 11) in both years. As soils dried to a tension of between 40-50cbars, irrigation was used. Regardless, there were differences in soil water tension due to ground cover but not nutrient source treatments. Across both seasons, the MB ground cover systems had the greatest soil water tension (drier soils) while the SP had the lowest soil water tension. It is believed that due to both the undertree vegetation of the MB system and the exposed soil as compared to the thickly mulched WC, GC, and SP treatments, soils were drier. However, soils were never at a level of water tension – so dry – as to be thought to impose growth effects. Conversely, with the SP treatments, soils were both measurably and noticeably wet. The SP created a machete over the soil. It was observed but not measured that soils smelled of anaerobic conditions. This observed effect usually dissipated by late summer or early fall as the machete started to breakdown.

Orchard pests occurrence
There were minimal pest problems in the newly established orchard during the first two years. There were only two insect pests. Damage by Japanese beetles (JB) was kept to a minimum by mass trapping 34,591 (in 2006) and 8,882 (in 2007) JB per trap by placing on 1 June 14 JB traps each baited with dual lures (floral + sex pheromone) spaced 30 m apart along east-west transect north of the organic fruit block. There was an infestation of JB feeding on trees in 2006. Trees with the MB ground cover sustained significantly greater foliar feeding damage than all other ground cover treatments, followed by the GC treatments (data not presented). JB feeding damage was correlated to the density of undertree vegetation. There was no effect of nutrient source on the JB feeding in 2006. In 2007, severe late summer defoliation by larvae of the leaf crumpler, Acrobasis indigenella (Zeller) was observed. This damage was restricted to the stunted and very unhealthy apple trees and occurred only in trees with the SP ground cover treatment. No other pests were evident due to lack of fruit. There was no significant incidence of diseases observed during the first two growing seasons. However, fireblight did occur on a few random trees and no treatment effects were observed.

Weed density varied with ground cover and nutrient source treatments. The MB had the greatest weed density or undertree vegetation density throughout the growing, followed by the GC treated trees. Application of either CF or PL nutrient sources increased vegetation density. It was noted that weeds were introduced into the system with the GC treatments. The WC trees had an intermediate, albeit relatively low undertree vegetation and weed density and the SP treated trees had significantly lower vegetation and weed density.

During the winter of 2007, field voles were observed in the orchard. Sticky traps were used to survey populations and trap-remove the pests. However, only a limited number of animals were trapped. This form of trapping was not repeated in 2008. However, in the winter of 2008 with significant snow and ice events, vole damage to the trees was observed and was significant. Trees with the MB treatment had severe vole feeding damage ratings (2.2 of 0 – 5 scale, 0 no damage, 5 = complete girdling; 2.5 = 50-70% girdling) and approximately 30% of the experimental trees were completely girdled. By comparison, the SP and WC trees had a damage ratings of 0.8 and 0.4 with no trees completely girdled, significantly less than the MB trees. There was not statistically significant effect of nutrient source on vole feeding damage, with the NF, CF, and PL trees having damage ratings of 1.0, 1.2, and 1.4, respectively. It may appear that the there was a slight increase, although not significant, in trees receiving nutrient treatments and this may be related to tissue nutrient content.

Economic Analysis
An economic analyses of the orchard is on-going. As part of the development of the project, and outreach, an interactive spreadsheet for economic analysis for this project has been developed. After the first two years, there were no significant differences in the total costs of operation although treatments varied with labor and supplies. With the first cropping year in 2008, year three of the orchard, the analyses of return on investments and net present value of the operation, as affected by treatment was calculated. The spreadsheet may be viewed at our project website,

Information dissemination
Information was presented to growers at regional grower meetings of the Arkansas-Oklahoma Horticulture Industries Show (Rom, 2007a; Rom, 2007b; Rom and McAfee, 2008a; Rom and Friedrich, 2005; Rom and McAfee, 2008b), other grower and trade groups (Rom, 2008a; Rom, 2008b; Rom, 2008c), at science and technical meetings (Choi and Rom, 2007; Choi and Rom, 2008; Rom, et al., 2008a) and the national SARE New Farm Conference (Rom, et al., 2008b).

Organic fruit workshop and field day An organic workshop and field day to disseminate research and production information about organic fruit production for the region was conducted at the UA was 9-November, 2007. A press release was sent to Arkansas, Oklahoma and Missouri Cooperative Extension Services. More than 110 persons participated, comprised mostly of Extension agents and market growers. The program provided talks on sustainable production systems including organic certification, organic nutrient management, sustainable soil and pest management. The results of SARE projects including the SARE Planning Grant, and this grant were reviewed, as well as other on-going organic fruit research at the UA. Research field tours of the replicated organic apple orchard and organic high-tunnel blackberries and raspberries were included. Best management practices for organic orchard establishment were explained and demonstrated. Each participant received a 3-ring binder with information covered in the workshop. All of the presentations were included on the UA Ecological Agriculture website,

This project has resulted in ten grower or trade group presentation, and six grower or trade publications. The project has resulted in four scientific or professional meeting presentations, three published abstracts and one published paper.

The project has resulted in three intitutional press releases with numerous publica media publications from releases.

Participation Summary

Project Outcomes

Project outcomes:

This project has been a very high profile project for our department and college. It has had numerous visitors and was the focus of the “Organic Fruit Production” workshop (described above). It has been the focus of several media reports. The most noticeable impact is the frequency with which the lead PI (Rom) has request for grower speaking engagements to present and discuss the preliminary project results. Likewise, there have been a significant number of inquiries by email and phone about organic apple production practices. At this time, it is difficult to quantify the use and application of the results of the project.

This project was envisioned and developed as a minimal 10-year long study. The current season (2008) is the third year of orchard growth and will be the first year of production. These results are expected to be more compelling for growers to adopt than just the early tree growth. It is expected that this project will be, not only an important source of science based information for organic production in the region, but an important demonstration tool as the orchard matures and enters its cycle of full production.

Based upon the development of this project and the growth, physiology, soil and soil biology, economic framework, and demonstration/teaching potential of this project, the PIs were able to solicit grant support from the USDA Integrated Organic Program to continue the project for an additional four seasons with the chance for renewal. The first grant cycle will take the orchard through its second growth phase – early production, and into its mature production phase. Thus, the SARE project was leveraged into a 4-fold support in resources and time.

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