A multi-site greenhouse climate and control dataset from commercial facilities in Spain (2021-2026)

This article introduces a greenhouse climate and control dataset collected from three commercial facilities in Álava, Navarra, and Cartagena, Spain. The data cover six greenhouse compartments operating under different climatic conditions, crop production systems, greenhouse structures, and climate-control configurations. The monitoring periods vary by site and run from 2021 to 2026. The dataset combines indoor climate measurements, outdoor meteorological observations, climate-control setpoints, and actuator states recorded by commercial Nutricontrol monitoring and control systems. The available variables describe temperature, relative and absolute humidity, vapor pressure deficit, solar radiation, wind and rainfall conditions, along with the operation of heating, ventilation windows, thermal and shading screens, and air circulation systems. Approximately 4.8 million one-minute timestamps were collected and then processed into about 319,000 observations at 15-minute resolution. Data curation included harmonization of variable names across facilities, variable-specific temporal aggregation, assessment and consolidation of redundant sensor streams, physical range checks, and robust outlier detection. Values removed during quality control remain explicitly represented as missing data rather than being generally interpolated or imputed. The resulting dataset provides real-world commercial greenhouse data suitable for climate modeling and forecasting, anomaly detection, control-system analysis, and the development and evaluation of data-driven control and digital-twin approaches.