Written by Srikanth Gorthi, Dattatray Bhalekar, Juan Munguia de la Cruz, Lav Khot, and Bernardita Sallato. Washington State University, June 2025
WSU Smart Apple Orchard: “Technology Testbed and Demonstration Site” was established in commercial WA38 apple orchard at Mattawa, WA. In 2024, this orchard was managed by Northwest Farm Management (NWFM) LLC, Yakima, WA. A commercial precision and automated irrigation system was established by three vendors, detailed below and at website (henceforth referred as precision block 1), and contrasted with the grower standard approach, based on soil moisture scheduling (henceforth referred as soil moisture block 2). Here we report on key findings for the first year of the study.
Key findings
- Precision irrigation system saved 52.4% of water when compared to soil moisture scheduling.
- Soil moisture scheduling led to larger fruit equivalent to one box size different (56 to 64 mm).
- Precision irrigation reduced green spot incidence to 6% compared to 23% in block 2.
- Overall, precision irrigation had higher packouts, estimated at 84%, compared to 45% in block 2.
- The estimated effective yield under precision irrigation 21% higher, with 232% higher water use efficiency compared to soil moisture scheduling.
- While the precision irrigation system utilized 52.4% less water, compared to soil moisture based irrigation, there were no signs of stress in the trees.
Smart Apple Orchard Testbed
WSU Smart Apple Orchard Testbed was established in WA38 apple orchard at Mattawa, WA, managed by NWFM, LLC, Yakima, WA (Cooperator: Mr. Keith Veselka and Carmelo Garcia). This is a four-year-old ‘WA 38’ orchard, trained as tall spindle, at 12 ft by 1.5 ft spacing (Figure 1). The first commercial harvest was in the year 2024, Year 1 of the testbed.
Irrigation Management
The orchard uses drip and sprinklers irrigation with automatic control. We evaluated two irrigation scheduling strategies; block 1 with precision automated irrigation based on soil, environment and plant base sensors, and block 2 with automation scheduling based on soil moisture. The precision system was automated using a centralized controller (RF-X1, Drop Control, WiseConn Engineering), with irrigation scheduling recommendations provided by SWAN Systems (SWAN Systems Inc.). SWAN Systems integrates data from multiple sources—including in-field soil moisture sensors (Drill & Drop 36”, Sentek Inc.), AWN weather station data, Planet satellite imagery, customized crop coefficients, and site-specific weather forecasts—to generate irrigation schedules. These schedules are then seamlessly integrated into the WiseConn Drop Controller for automated irrigation. Block 2, the automation also used a centralize controller and the irrigation was scheduled based on soil moisture probes data, maintaining the soil moisture between 100 and 75% field capacity. These probes are installed at 8, 16, 24, and 36 inches below surface.
Dynamax Inc. also deployed sensors to monitor sap flow (Sap flow SGEX-25), growth/shrinkage of trees and fruit (DEX-70 Dendrometers), soil moisture (SM150 Delta-T), and crop water stress (SapIP-IRT sensors). These sensors were deployed in both blocks.
Ground-truthing/Evaluations
We monitored multiple aspects of soil, plant and environment throughout the season. These included water flow meters (SW-3L, Dragino Inc.) to monitor and quantify irrigation water applied, soil moisture sensors at 8 and 24 in using (TEROS 12, Meter Group Inc.) and weather using an all-in-one weather station (ATMOS 41, Meter Group Inc.) installed inside and outside the orchard. Stem water potential was determined using micro-tensiometers (MT FloraPulse, Co.) inserted into the trunks for real-time plant water status monitoring. All the data was collected through LoRaWAN (Long range wide area network) based wireless sensing network (WSN), as part of WSU AgWeatherNet Smart Farm platform (for more details please click here).
In addition to sensing, shoot and fruit growth was monitored throughout the season and at commercial harvest, ten trees from four random locations (two per block) (Figure 1), were harvested to analyze total fruit per tree, fruit defects, fruit diameter, weight and firmness. With the information collected, we estimated fruit yield, packouts, water use efficiency, and economic benefit on a per acre basis.
Irrigation Management Comparison
The precision irrigation system in block 1 had 52.4% less water applied compared to block 2 (Figure 2). SWAN Systems model driven irrigation scheduling applied 8.28 inches of water during the growing season (July 1 to October 15) compared to 15.8 inches in block 2 for the same period.
Yield and Fruit Quality
Block 1 had 9.4% less fruit per tree and small sized fruit (Table 1), with 69% of the fruit within 79-91 mm diameter (box size 113 to 56) (Figure 3). However, fruit defects were 64.4% lower in block 1 compared to block 2 with sunburn (necrosis) and green spot (Table 2) being major contributors to those losses. Higher green spot in block 2 could be attributed to higher vigor induced by higher irrigation (for more information read this summary report).
Table 1. Fruit quality and productivity parameters collected at harvest.
| Site | Fruit count (n) | Fruit Weight (g) | Diameter (mm) | Firmness (Lb) | Fruit 79 – 91 mm (%) |
|---|---|---|---|---|---|
| Block 1 | 19.1 | 280b | 81b | 20.8 | 69b |
| Block 2 | 21.1 | 318a | 84a | 20.9 | 88a |
| p value | 0.71 | 0.01 | 0.05 | ns | 0.05 |
*Different letters for each quality attribute (column) indicate significant difference. ns indicates not significant difference at p < 0.05
Table 2. Fruit defect parameters collected at harvest.
| Site | Necrosis (%) | Green Spot (%) | Photooxidative (%) | Browning (%) | Total Defects (%) |
|---|---|---|---|---|---|
| Block1 | 4b | 6b | 0 | 0 | 16b |
| Block2 | 13a | 23a | 2 | 1 | 45a |
| p value | 0.037 | 0.002 | ns | ns | <0.001 |
*Different letters for each quality attribute (column) indicate significant differences. ns indicates not significant difference at p < 0.05
Plant Available Water and Soil Moisture Analysis
Plant available water (PAW) was calculated from soil moisture data with a permanent wilting point set at 10% and a field capacity at 32%, based on loamy sand texture. SWAN Systems driven irrigation scheduling maintained PAW within the 50–75% range, while the soil moisture driven schedule consistently replenished water up to field capacity (Figure 4). Notably, during the month of July, Site 2 had significant decline in PAW, compared with Site 1, both irrigated equally. This difference might be associated to differences in the soil profile and effective depth observed on site (data not shown here).
Stem water potential (SWP) ranged between -1 and -10 bars in block 1, while in block 2, site 4 reached levels below -10 bars, suggesting water stress (Figure 5), despite receiving the same amount of water. Higher water stress in site 4 aligns with the soil moisture content described above, in which the south side of the orchard had also reduced PAW at 24” depth (Figure 4), associated to shallower effective soil depth. Overall, Sites 3 and 4 exhibited comparable mid-day SWP to that of Sites 1 and 2.
This results also suggests that relying on a single measurement approach (e.g., soil moisture or environmental conditions) may not adequately capture plant water demand. Instead, integrating multiple sources of information, including soil, plant and environmental indicators provide a more comprehensive foundation for effective irrigation management.
Cost-Benefit Analysis
Precision irrigation system utilizing soil, plant and environmental data, here analyzed and interpreted by SWAN Systems, tied to an automated system (WiseConn), reduced water consumption by 52.4%, while increasing fruit quality and yield by 21%. This translates to 232% higher water use efficiency, as a measure of crop per drop. The estimated water use efficiency in lbs of marketable fruit per thousand gallons was 116 lbs in block 1, compared to 50 lbs block 2. Considering an average of $35 USD per 40 lbs box in WA-38 apples (Gallardo et al., 2025), the precision irrigation in block 1 had $4,084 USD/acre increase returns. We will continue evaluating these systems in 2025.
Acknowledgements
This project is funded by the Washington Tree Fruit Research Commission. We thank all the private partners who have deployed their technology, as in-kind. The authors acknowledge Dr. R. Troy Peters, Ms. Prasanna Medarametla from WSU and specially Keith Veselka, Carmelo Garcia and Miguel Sanchez from Northwest Farm Management for their help in the completion of this study.
Additional Readings
Gallardo, K. Galinato, S and B. Sallato. 2024. 2024 Cost and Return Estimates of Establishing, Producing, and Packing Cosmic Crisp Apples in Washington. Washington State University Extension Publications. TB104E.
Sallato, B and L. Khot. 2024. Soil and Plant Diagnostic Technology for Orchard Nutrient Management. Fruit Matters summary report. https://treefruit.wsu.edu/article/soil-and-plant-diagnostic-technology-for-orchard-nutrient-management/
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.
