Guang Yi, Xilin Wang, Songyu Jiang, Jianfeng Zhao, Tengfei He, Zhaohui Chen
Continuous multi-parameter monitoring is important for characterizing environmental conditions in commercial beef cattle facilities, but sensor performance and system reliability require evaluation under production conditions. This study developed a LoRa-based multi-sensor node for air temperature, relative humidity, CO2, NH3, air speed, and illuminance. Field performance was evaluated at one commercial farm by synchronous comparison with commercial instruments using 1-min paired observations and regression- and agreement-based analyses. Nine prototype nodes were subsequently deployed at three sites for 14-day monitoring periods. Temperature and relative humidity showed the strongest linear relationships with the comparison instruments (R2 = 0.995 and 0.994), followed by illuminance, air speed, and CO2 (R2 = 0.990, 0.863, and 0.773). NH3 showed a weaker relationship (R2 = 0.604), improving after 10-min aggregation; most application-monitoring estimates below 5 ppm were extrapolated rather than validated absolute concentrations. Overall data availability was 99.66%, and continuous monitoring captured temporal patterns in thermal conditions, air quality, local airflow, and illuminance. The evaluation focused on environmental monitoring performance rather than animal-based outcomes. These findings support parameter-specific environmental monitoring and multi-site deployment in commercial beef cattle facilities, with potential for future anomaly detection and environmental control.