Skip to main content Skip to navigation

Soil and Plant Diagnostic Technology for Orchard Nutrient Management

Written by Bernardita Sallato and Lav Khot, January 2025.

Effective nutrient management in tree fruit production requires a comprehensive understanding of multiple interacting factors, including soil conditions (physical, chemical, and biological), plant health and its nutrient demand through various growth stages, water availability, and sub-seasonal weather conditions. Current methods like soil and leaf tissue testing, standardized by the Soil Science Society of America (SSSA), remain underutilized by growers due to a perceived lack of correlation between test results and tree expressed outcomes in growth, yield, and fruit quality. Also, growers and service providers are looking for ew rapid and potentially non-destructive technologies to better estimate soil and plant nutrient status for informed management decision making.

In pursuit of such technologies, we evaluated new and existing nutrient management tools, across two cultivars and sites: A commercial ‘Honeycrisp’ orchard block near Grandview, WA (2023 season), and a ‘WA 38’ orchard block near Mattawa, WA (2024 season, Figure 1). Explored technologies include  : macro-/micro-nutrients mapping by SoilOptix, Electric Conductivity (EC) by Simplot, Web Soil Survey as well as remote sensing (satellite and drone) RGB and multi-spectral imagery data products. Site-specific data at various spatiotemporal scales was correlated with ground truth soil and plant nutrient levels as well as with productivity parameters (e.g., tree and fruit growth, fruit quality). Two alternative plant nutrient monitoring tools—Sap analysis by Advancing Eco Ag and LAQUAtwin (Horiba)—were also tested in this project. Below is a brief overview of technology and pertinent research outcomes.

"aereal map of orchards with locations"Figure 1. Site selection in WA 38 block (2024 season), with planting density and irrigation differences highlighted in colored boxes: light green (left) for a planting distance of 3 ft by 11 ft, and double (2x) the irrigation, dark green (middle) for a planting distance of 1.5 ft by 11 ft and double (2x) the irrigation, and blue (right) for a planting distance of 1.5 ft by 11 ft and regular irrigation (1x).

Specific Findings 

  1. Soil Mapping and Testing:
    • SoilOptix® mapping showed moderate accuracy for some nutrients (calcium and sulfur) but this sensing module needs site-specific calibration.
    • Standard soil tests revealed nutrient deficiencies, particularly for potassium and calcium.
    • Soil nutrient differences affected tree performance, with higher productivity in areas with balanced nutrient levels.
  2. Plant Nutrient Testing and Fruit Quality:
    • Optimal conditions (e.g., balanced potassium and lower nitrogen) produced better-quality apples with fewer defects in WA 38. High nitrogen levels increased fruit defects (e.g., green spot). Excess nitrogen also reduced fruit weight and overall productivity.
    • Potassium showed a positive correlation with fruit size, weight, and productivity.
    • Sap analysis data correlated with nutrient levels and fruit characteristics, but results varied across sites.
    • Portable LAQUAtwin testing kits were unreliable for accurate nutrient testing.
  3. Technology and Data Use:
    • Aerial mapping tools helped identify areas for targeted sampling but could not independently predict nutrient levels reliably.
    • Vegetation indices, like NDVI and NDRE, correlated with plant nutrient levels and provided insight into tree health. This aspect is being explored further with our efforts at Smart Orchard testbed.

Results Summary

Soil nutrients mapping

SoilOptix® uses gamma radiation-based sensor for high-throughput mapping of physical and micro-/macro- nutrients in the topsoil. Though useful, this sensing module requires additional calibration for reliable nutrient recommendations It demonstrated weak or no correlations with actual measurements in ‘WA 38’ cultivar, while in the ‘Honeycrisp’ orchard there was a strong correlation with calcium (Ca, r = 0.96), magnesium (Mg, r = 0.74) and boron (B, r = 0.93).

The soil E.C. mapping provided close correlation with Na.

WebSoil Survey (USDA) application has the largest amount of soil characteristic information such as soil chemical, physical, environmental and health indicators. However, this information is available at a macro scale and requires calibration. Other tools such as Google Earth, and drone imagery also aided in guiding on the ground sampling decisions. More details on the mapping technology for soil variability can be found in our previous summary Mapping orchard variability and soil attributes to improve site-specific management decision making and also in Sallato (2004).

Leaf nutrients mapping

Ground-truth leaf nutrient levels showed correlations with productivity metrics like fruit count, weight, yield, and disorders. However, these correlations varied by cultivar and site. In Honeycrisp, low nitrogen (N, %) was linked to smaller fruit size. In WA 38, higher N levels were positively correlated with green spot and shoot growth, and closely related to the irrigation strategy and vigor. Leaf N was negatively correlated to total fruit yield (r = -0.70) and count (i.e., crop load, r = -0.69), while these indicators were positively correlated with leaf potassium (K, %)(Figure 2).

"correlation charts"

Figure 2. Correlation between total fruit weight per tree (Kg) (Top) and fruit per tree (bottom) with leaf N (left) and leaf K (right) for WA 38 variety.

In WA 38, green spot incidence on harvested fruit varied between 0 to 35%. This ground-truth data showed exponential model fit with leaf N (Figure 4). Note that leaf N is not the only contributing factor as higher green spot incidence was observed in the areas with higher shoot growth, higher leaf N levels, and higher nitrate (NO3) levels in the soil. This high crop vigor was related to increased irrigation (50% more) and may have combined effect with higher water retention in the soil, compared to the balanced areas.

"exponential chart"

Figure 3. Exponential relation between green spot incidence and leaf N. The dotted line indicates 90% probability, strait gray line indicates 95% probability.

The 5-band multi-spectral (sensor: RedEdge3, Micasense Inc., WA) aerial imaging data revealed useful correlations with tree nutrient levels. Positive correlations were observed between leaf N and Normalized Difference Red Edge Index (NDRE), and Ca with Normalized Difference Vegetation Index (NDVI, Figure 5). Similarly, negative correlations were observed between iron (Fe,%) with Renormalized Difference Vegetation Index (RDVI) and Soil Adjusted Vegetation Index (SAVI). These findings suggest that, with additional sites and validation efforts, aerial multispectral imagery data could aid in creating micronutrients prescription maps needed to support variable-rate nutrient application. This is the focus of our new project “New Strategies for Precision Plant Nutrient Application” funded by the Washington Tree Fruit Research Commission.

"aereal map of orchard"

Figure 4. Drone imagery based Normalized Difference Vegetation Index map of the Honeycrisp block near Grandview, WA (Aug 9, 2023).

Alternative nutrient (e.g., Sap analysis) estimation methods showed varied levels of correlations across the cultivars and sites. In Honeycrisp cultivar, magnesium (Mg), ammonium (NH4), and molybdenum (Mo) levels in the sap correlated with crop load. Similarly, N, P, and Ca in the sap were positively and moderately correlated with fruit size. In WA38, K in the sap had a strong and positive correlation with fruit count and yield. Also, the areas that were identified as Ca deficient by Sap analysis were also low in soil Ca levels. We did not implement nutrient management recommendations based on Sap results which would be key to determine the cost-benefit of this method.

The LAQUAtwin, that uses ion selective substrates, did not yield reliable results for nutrient diagnostics.

 Recommendations

  1. Soil Management:
    • Regular soil testing and mapping are essential to identify and address nutrient imbalances.
    • Focus on maintaining balanced potassium and calcium levels while avoiding excessive nitrogen.
  2. Plant Nutrition:
    • Monitor leaf nutrient levels to guide fertilization strategies.
    • Avoid over-fertilizing with nitrogen to minimize fruit defects and ensure better fruit quality.
  3. Technology Utilization:
    • Use multispectral aerial imaging derived vegetation indices to monitor tree health and identify problem areas.
  4. Limitations of Portable Tools:
    • Portable testing kits (e.g., LAQUAtwin) provided inconsistent results in our resarch trails.

 The Smart Apple Orchard project will continue evaluating some of the existing and emerging technologies and delivering pertinent education to the grower community. For more information contact our team:

Contact

Bernardita Sallato professional photo

Bernardita Sallato

b.sallato@wsu.edu

Lav Khot professional photo

Lav Khot

lav.khot@wsu.edu

Acknowledgment

This project was funded by WTFRC Technology Committee. We would like to thank our grower cooperators (Washington Fruit & Produce and North West Farm Management LLC), as well as WSU graduate students (Elda Bezuayene, Gajanan Kothawade, Nipun Thennakoon, Juan Carlos Munguia de la Cruz) who helped with the research efforts.

References

 

Fruit Matters articles may only be republished with prior author permission © Washington State University. Reprint articles with permission must include: Originally published by Washington State Tree Fruit Extension Fruit Matters at treefruit.wsu.edu and a link to the original article.

Washington State University