Written by Srikanth Gorthi and Lav Khot, AgWeatherNet, Washington State University, April 4, 2025
Overview of AWN Tier-3 Program
WSU AgWeatherNetwork (AWN) currently offer growers to add their private weather stations into AWN ecosystem (more details at: AWN Tier-3 Stations). AWN does real-time data pull from these stations, at 15-min interval, quality checks, and alert growers about any issues with the weather station data or maintenance needs. These stations are also tied to the decision support tools available within AWN and Decision Aid System. We have now expanded this ecosystem to accept block level distributed sensors to realize private sensing networks. This ecosystem enables real-time data-driven decision making for effective frost, heat stress, and irrigation management. Growers can integrate their preferred low-cost sensor deployments—covering weather, plant, and soil conditions—into a broader open-field weather monitoring ecosystem maintained by AWN. As these private sensing networks are driven by Long Range Wide Area Network (LoRaWAN, detailed in next section), it reduces overall data collection costs, compared to private industry offered subscription driven solutions. Additional benefits of this ecosystem include scalability, long-term and secure data storage, and data privacy.
Private Sensing Network
The AWN Private Sensing Network, i.e., AWN Smart Farm(s), primarily uses LoRaWAN compatible sensors for monitoring and managing the orchard block. LoRaWAN is particularly well-suited for agricultural applications, needing time-series data, due to its low-bandwidth long-range communication (up to 10 miles between gateway and sensor nodes) capability, low power consumption, and minimal hardware maintenance costs. For more details about LoRaWAN technology, read Pagano et al. (2023).
The AWN Smart Farm ecosystem is compatible with most LoRaWAN-based sensors. A wide range of commercial LoRaWAN-based sensors are available for soil, plant, and weather monitoring, and can be procured directly from private vendors. Examples include S31-LB/LS, D20/D20S-LB, SE0X-LB, SenseCAP S2120, SenseCAP S2101, RAKWireless RAK1901. In addition, sensors that communicate using the serial data interface (SDI-12) can also be integrated into LoRaWAN networks via compatible dataloggers. Common SDI-12 sensors include the ATMOS-14, TEROS-12, Florapulse SWP, LI-710 and can be connected to LoRaWAN dataloggers (nodes) such as Dragino SDI-12. Each individual sensor (/node) transmits the data to a field-deployed gateway (located within 10 mile radius). Gateway serves as a communication bridge, via on site WiFi or cellular connection, to the remote AWN cloud (Figure 1). To ensure wide and reliable coverage, gateway antennas should be installed at elevated locations, for example on equipment storage sheds to avoid obstructions between nodes and gateway. AWN is currently conducting a comprehensive research study to evaluate various compatible sensors for their accuracy, compared to scientific grade weather sensors, as well as robustness of gateways in terms of reliable data collection and transmission.
A typical private sensing network data not only helps in monitoring, and passive decision making but also powers automation. Based on sensor inputs and predefined thresholds or models, triggers can be sent to actuate overhead evaporative cooling, irrigation systems, and wind machines.
AWN Smart Farm Dashboard
The AWN Smart Farm Dashboard delivers real-time data from each of the private sensing network. The dashboard provides data insights for individual orchard or vineyard block(s). This data is only accessible to specific user(s), and at varied admin levels, defined by the owner of the private sensing network. Figure 2 illustrates an example of a private sensing network deployed in a commercial sweet cherry orchard to monitor frost conditions. In this deployment, data from air temperature, radiative frost, and bud tissue temperature (measured using thermistor inserts into buds) sensors are being collected and pushed to the dashboard at 1-min interval. In addition to in-field sensor data, the dashboard renders data from the nearest (grower preferred) AWN open-field weather station. This enables a more accurate assessment of frost events and supports intelligent decision to proactively manage potential bud damage. Growers can also use open-field weather station data in identifying anomalies within the private sensing network or vice versa.
Use cases and future
Besides frost monitoring, we have deployed pilot private sensing networks in our grower cooperator blocks (blueberry), and WSU Smart Apple Orchard and DEMO Smart Vineyard testbeds. We are monitoring frost (blueberry), heat stress (apple, grapes) conditions and have automated heat stress mitigation using localized sensor inputs. These pilots are also monitoring soil moisture, temperature, and overall block specific water use through compatible flow meters to help us compare water use in irrigation and heat stress management events. In near future, AWN Smart Farm Dashboard will be customized to visualize crop phenology (e.g., Growing Degree Days), and crop/cultivar specific decision support models (e.g., cold hardiness, heat stress), weather guided optimal spray-time windows for efficient crop protection, among others. As research progresses, we also plan to integrate localized nowcasts for enhancing advanced warning capability for block specific management decision making.
If you want to add such private network or have feedback, please reach out to us at weather@wsu.edu
Acknowledgments
This product is an outcome of PhD dissertation work done by Srikanth Gorthi (Advisor: Lav Khot) and was supported in parts by NSF/USDA NIFA Cyber-Physical Systems (Award Nos: 2021-67021-34336, 2021-67021-35344 [AgAID Institute]), USDA NIFA 0745 projects, and AWS Cloud Credit for Research. We also want to acknowledge Sean Hill and Sanjita Bhavirisetty at AgWeatherNet for their help in porting, scaling, and improving the AWN web ecosystem. The authors also thank our grower cooperators (Olsen Farms, Lighthouse Farms, and Hayden Farms, NWFM LLC.) and WSU collaborators (Dr. Markus Keller, Dr. R. Troy Peters, Prof. Gwen-Alyn Hoheisel, Associate Prof. Bernardita Sallato).
Additional Reading
Pagano, A., Croce, D., Tinnirello, I., & Vitale, G. (2023). A Survey on LoRa for Smart Agriculture: Current Trends and Future Perspectives. IEEE Internet of Things Journal, 10 (4), 3664–3679. https://doi.org/10.1109/jiot.2022.3230505
Contact
Lav Khot
Director of AWN
lav.khot@wsu.edu
(509) 786-9302
(509) 335-5638
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.
