Wireless Sensor Network: New Concept of Spatial-Temporal Monitoring Plant-Environment Interactions

We present a low-cost, standards-based wireless sensor network (WSN) for continuous, canopy-integrated monitoring of plant-environment interactions. Each plant carries in-canopy microclimate sensors (temperature, relative humidity, illuminance) paired with nearby ambient references, yielding real-time canopy-ambient differentials. The system is easy to install: at planting or sowing, sensors are fixed at positions that will lie within the developing canopy, and a separate ambient reference area is designated and kept free of vegetation. As plants grow, they envelop the sensors, thereby capturing growth dynamics over time. The sensors accuracy was validated against a commercial weather station and portable system that measures gas exchange, temperature and light (LI-COR 6800/6400), and the systems ability to resolve plant physiological activity was confirmed using the PlantArray functional phenotyping platform with independent whole-plant transpiration and biomass references. Under controlled growth-room conditions and across two contrasting Cannabis cultivars, daily transpiration strongly predicted biomass gain (R{superscript 2} > 0.9). Microclimate signals mirrored physiology: midday canopy air was cooler by 4-7 {degrees}C, more humid by 18-25 % RH, and increasingly shaded as biomass accumulated, with temperature, RH, and light attenuation showing saturating logarithmic relationships with growth. The network operated for months unattended with low packet loss and predictable power use. It provides 4D (x-y-z-time) coverage, where x and y denote horizontal location, z the vertical position within the canopy, and time the dynamics, enabling resolution of where changes occur and how they evolve, and supplying high-frequency labeled data. This system complements, rather than replaces, precision instruments and high-end phenotyping platforms, providing a scalable layer for continuous tracking across wide areas. We outline practical constraints and next steps toward field pilots, modest energy harvesting, expanded sensor suites, and integration with machine learning for predictive crop management.